Journal of Cleaner Production 419 (2023) 137871

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Real-time energy flow mapping: A VSM-based proposal for

energy efficiency
Leonardo Fontoura a, Daniel Luiz de Mattos Nascimento b, *, Julio Vieira Neto a,
Noelia Garcia-Buendia c, Jose Arturo Garza-Reyes d, Gilson Brito Alves Lima a,
Guilherme Luz Tortorella e, Rodrigo Goyannes Gusmão Caiado f, Marcelo Jasmim Meiriño a
a
Fluminense Federal University, Niterói, Brazil
b
University of Barcelona Business School, Barcelona, Spain
c
Department of Business Organization, Marketing, and Sociology, University of Jaen, Jaen, Spain
d
Centre for Supply Chain Improvement, University of Derby, UK
e
Mechanical Engineering Department, The University of Melbourne, Melbourne, Australia
f
Pontifical Catholic University of Rio de Janeiro, Rio de Janeiro, Brazil

A R T I C L E I N F O A B S T R A C T

Handling Editor: Kathleen Aviso Rising energy costs and the current concern about sustainability issues directly impact the competitiveness and
survival of organizations, promoting the search for continuous improvement in industrial production processes.
Keywords: The present study, by identifying the absence in the literature of models based on Value Stream Mapping and
Lean management integrated with Industry 4.0 technologies that promote better energy efficiency practices in industrial processes,
Value stream mapping
proposes a new Lean Management model called Real-time Energy Flow Mapping. This innovative model, provide
Industry 4.0
capabilities to a visual, digital and sustainable management system of all energy behavior in machines and
Energy efficiency
Delphi study equipment using fully parameterizable metrics and rules. Also, estimulating stakeholders with the ability to
Fuzzy logic predict and monitor energy costs, as well as identify possible problems due to changes in the behavior of energy
consumption in machines, thus helping managers to make decisions faster and more dynamically. As a result, the
literature search turned up 17 Value Stream Mapping based models with a focus on energy efficiency, along with
31 metrics grouped on energy, environmental and production processes. After discussions by the Focus Groups,
16 resulting metrics were submitted to expert panel, in which the least relevant were discarded and the 13 most
relevant classified using Fuzzy Delphi. An analysis to verify the interrelationships between the metrics is per­
formed using the Fuzzy DEMATEL, being finally applied in the proposed model. A case study of the model was
carried out, submitting it operationally at the LabFaber 4.0, empirically validating the study. The Real Time
Energy Flow Map fills the gap in the literature on Value Stream Mapping models to improve energy efficiency,
monitoring energy consumption through dynamic mapping of the energy flow of production processes, identi­
fying the area with increased potential for energy savings through changes in operating behavior that reduce
overall energy consumption and environmental impacts.

worldwide electricity consumption in just under 50 years (IEA, 2021b).

Consequently, ongoing global concerns around energy consumption and
1. Introduction climate change and the quest for more efficient energy use and man­
agement have led to the concept of sustainable development becoming
The 1973 oil crisis triggered the realisation that energy raw materials consolidated in various countries (Bazurto et al., 2017). In addition to
are exhaustible and will become increasingly scarce over time (Ates and the concerns around efficient energy use, this concept also considers the
Durakbasa, 2012). With the globalisation that integrated nations into a cost of electricity (Akbaba, 1999), which has constantly increased in
world economic system (Chen et al., 2019), the development of coun­ multiple countries worldwide (Adenuga et al., 2019).
tries’ economies has been mirrored by an increase in energy demands (X. The industrial sector is responsible for approximately 37% of global
D. Wu et al., 2019), culminating in an over five-fold increase in

* Corresponding author.
E-mail address: [email protected] (D.L.M. Nascimento).

https://doi.org/10.1016/j.jclepro.2023.137871
Received 7 February 2023; Received in revised form 18 June 2023; Accepted 19 June 2023
Available online 10 July 2023
0959-6526/© 2023 Published by Elsevier Ltd.
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Abbreviation LCAVSM Life Cycle assessment Value Stream Mapping

LM Lean Management
BDA Big-Data Analytics MFCA Material Flow Cost Accounting
CO Carbon Monoxide NH3 Ammonia
CO2 Carbon Dioxide NO2 Nitrogen Dioxide
CPS Cyber-Physical-Systems OT Operational Technology
DVSM Dynamic Value Stream Mapping RQ Research Question
ESG Environmental, Social and Corporate Governance RTEFM Real-time Energy Flow Mapping
EVSM Energy Value Stream Mapping SO2 Sulfur Dioxide
H2S Hydrogen Sulfide SVSM Sustainable Value Stream Mapping
I4.0 Industry 4.0 TFN Triangular Fuzzy Numbers
IEVA Integrated Energy Value Analysis TVSM Therblig-embedded VSM
IoT Internet of Things VSM Value Stream Mapping
IPS Interpretive Structural Modelling VVLD Value - Value Load Diagram
IT Information Technology WoS Web of Science

energy consumption (IEA, 2020) and 23% of CO2 emissions (IEA, 2022) and productivity enhancement (Tripathi et al., 2021), among
2021a). Economic crises and the strong competitive environment have others.
prompted organizations to seek better commercial and strategic op­ As can be observed, several works have addressed the use of VSM for
portunities to boost their profits and market share (Valamede and assessing energy efficiency issues (Cosgrove et al., 2018) or used a LM
Akkari, 2020). In this context, as energy-related issues are extremely perspective in combination with I4.0 technologies (Pagliosa et al.,
important for companies in the industrial sector, energy-efficient pro­ 2021). Traditional VSM neglected sustainability aspects as no energy
duction processes can guarantee a firm a market competitive advantage. indicators or metrics were included (Phuong and Guidat, 2018). The
The competitive and dynamic environment that companies are lack of models based on the VSM with a greater focus on sustainability
immersed in requires their constant adaption, with new practices, to (J. K. Y. Lee et al., 2021) has led to the need to carry out more advanced
meet market needs, while also keeping their processes sustainable (A.-Y. studies on the sustainable digital VSM topic (Jamil et al., 2020; Sulaiman
Chang and Cheng, 2019). An energy management system to improve et al., 2019). As it did not reflect the dynamic changes that occurred in
energy efficiency provides an organisation with guidelines to establish production (Balaji et al., 2020), it needed to be integrated with tech­
energy policies, execute efficient audits, calculate consumption trends nologies capable of obtaining data in real-time (Huang et al., 2019a).
and estimate and compare energy data, as well as support for This research seeks to investigate the following research questions
decision-making (Adenuga et al., 2019). (RQs):
Evidence suggests that Lean Management (LM) tools help organiza­
tions achieve better operational performance by fulfilling the key re­ RQ1. How can a LM model integrated with energy, environmental
quirements of profitability, quality and customer satisfaction (J A and production process metrics improve energy efficiency in
Garza-Reyes et al., 2018). LM practices used in organisational man­ industries?
agement include Just in Time (JIT), Total Productive Maintenance RQ2. What are the most relevant energy, environmental and pro­
(TPM), Jidoka, Value Stream Mapping (VSM) (Andreadis et al., 2017), duction process metrics that can be integrated into an energy man­
Kanban and Kaizen (Keskin et al., 2012). According to data from a agement model?
survey of specialists, one of the most popular practical tools is VSM with RQ3. What interrelationships exist between energy metrics on
74.3% preference (J A Garza-Reyes et al., 2018). VSM is a lean tool environmental and production process metrics?
based on visual presentation used to illustrate, identify and measure the
waste generated by inefficient processes (Abdulmalek and Rajgopal, The present study proposes and apply a VSM-based model named
2007). VSM has previously been used to address energy efficiency Real-time Energy Flow Mapping (RTEFM) that integrates energy, envi­
focusing on energy consumption (Sihag and Sangwan, 2019) and CO2 ronmental and production process metrics, and I4.0 technologies to
emissions (Heravi et al., 2020), life-cycle assessment (Estrada-Gonzalez automatically and dynamically collect, map and model the energy flow
et al., 2020), carbon efficiency and carbon emissions (Zhu et al., 2020) in an industrial production process to address the above-mentioned is­
and energy waste (Jia et al., 2017), among others. sues. From a theoretical perspective, this work contributes to advancing
The fourth industrial revolution (I4.0) is envisaged as the fusion of knowledge on the implications of LM practices for sustainability issues
information technology (IT) and operational technology (OT) (Ferreira such as energy efficiency and the advantages of I4.0 technologies. In
et al., 2022) and focuses on the end-to-end digitisation and integration practical terms, managers can adapt the proposed tool to the unique
of digital industrial ecosystems, mainly Cyber-Physical-Systems (CPS), features of their production systems and validate it to improve energy
Internet of Things (IoT) and Cloud Computing technologies (Xu et al., efficiency control, failure analysis, energy contracting budget fore­
2018). The emergence of I4.0 in parallel with the digital transformation casting and decision-making through an aggregated visual approach.
has made the technologies used in industrial production systems This research contributes to methodology by concisely integrating
increasingly smarter (Schumacher et al., 2020), which creates a major and applying several research methods. Starting with a qualitative
challenge to the identification and implementation of strategic methods analysis to better understand industries’ energy efficiency issues and
and the use of lean tools (Dombrowski and Mielke, 2015). I4.0 and LM needs, it develops a new energy management model through a scoping
are interrelated and I4.0 can enhance LM practices (Schumacher et al., review and focus group discussions among specialists. The study sub­
2020). The VSM approach has previously been studied in combination sequently uses Fuzzy Delphi and Fuzzy DEMATEL to quantitatively
with I4.0 technologies focused on the circular economy (Nascimento analyse the results to identify the most relevant energy, environmental
et al., 2022), a human-centred approach (Wang et al., 2022), and production process metrics and their strongest interrelationships.
manufacturing (Huang et al., 2019a), healthcare operations (Tortorella Finally, a case study using the model is applied in a real company,
et al., 2022), connectivity and information exchange (Nounou et al., LabFaber 4.0, to empirically test and validate its operation.

2
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

This document is organised as follows. This section seeks to con­ mapping material and information flow. The traditional VSM is a static
textualise and introduce the research needs, gaps, questions and moti­ tool (Balaji et al., 2020) and changes in the production line or products
vations. Section 2 presents a summarised background to the models can directly impact the flow mapping, making updating it a constant
found in the literature and the barriers to their implementation. Section challenge for specialists (Buer et al., 2018), therefore, integration with
3 explains the four-step methodology used in detail: scoping review, I4.0 technologies can improve the VSM by collecting data and process
focus groups, questionnaire design, two-round Delphi study using the information in real time, making the tool more dynamic (Meudt et al.,
Fuzzy Delphi and Fuzzy DEMATEL techniques, and case study. Section 4 2017). There are several technologies that are part of I4.0, as well as
presents the main study results following the structure of the RQs and different understandings about which technologies are part of this
the proposed conceptual model, while Section 5 discusses the main concept. I4.0 is supported by nine technological pillars: Big-Data,
research findings. Section 6 summarises the study and offers some Autonomous Industrial Robots, Simulation, Horizontal and Vertical
conclusions. Integration of Systems, IoT, Cyber-Security, Cloud Computing, Additive
Manufacturing, and Augmented Reality (Pereira and Simonetto, 2018).
2. Background In addition to these I4.0 technologies, Blockchain can be highlighted,
which today is the most critical and vital emerging computer technology
As a result of constant energy shortages and rising prices for energy to achieve energy efficiency in buildings. Whose three main factors that
supplied, improving energy efficiency in equipment contained in in­ made Blockchain an essential computer technology are real-time
dustrial processes has become one of the ways to increase the company’s execution, security and transparency, followed by traceability and
profits and reduce the emission of greenhouse gases. According to localized control (Qarnain et al., 2021).
Johansson and Thollander (2018), issues related to energy efficiency are In the literature, the year 2022 is characterized by a widespread
not something optional for companies, but extremely important for their energy crisis, which leads to a constant increase in electricity prices
survival, directly reflecting on the final price of the product. The LM worldwide (Saligkaras and Papageorgiou, 2022). This global energy
concept aims to ensure the company’s development and survival, being crisis stems from the Russian-Ukrainian conflict and has raised the
a business strategy focused on working with minimum resources to prospect of returning to some sources of electricity that are far from
achieve total quality (Valamede and Akkari, 2020). VSM, also known as being green and ecologically correct, thus forcing industries to adopt
material and information flow mapping, is a Lean tool used to analyse, immediately new initiatives to reduce energy consumption (Borowski,
examine, and evaluate specific work processes in a manufacturing 2022) as much as possible (Saktiawan et al., 2022). Futhermore, the lack
operation (Verma and Sharma, 2019), being used to map the flow of the of sustainability-oriented studies regarding VSM (J. K. Y. Lee et al.,
current process, helping specialists to identify areas where improve­ 2021), together with the lack of research on the integration and sys­
ments may occur (Buer et al., 2018). As shown in Fig. 1, flow mapping is tematization of available knowledge about VSM (Gholami et al., 2019),
one of the most popular Lean tools (Thanki and Thakkar, 2016) and is has led to a growing number of professionals and researchers to propose
considered one of the most efficient ways to obtain data from production the need to carry out more advanced studies concerning the
processes (Camgoz-Akdag et al., 2018). sustainability-oriented VSM (Jamil et al., 2020; Sulaiman et al., 2019),
Fig. 1 shows an example of a VSM applied in a production line, from given the difficulties in analyzing all relevant data from different
the entry of raw materials to the supply of final products to distributors, products, as well as the complexity of verifying essential information

Fig. 1. Example of a VSM applied to a production line (Ferreira et al., 2022).

3
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

related to environmental and production processes issues in a single of I4.0 technologies with VSM (Valamede and Akkari, 2020), in order to
VSM (J. K. Y. Lee et al., 2021). Added to this lack of studies, no previous improve the service quality, prioritizing the solution of the main prob­
studies have integrated energy efficiency and I4.0 technologies into a lems in the process (Camgoz-Akdag et al., 2018), not necessarily in
VSM-based approach to address dynamic changes and complex energy real-time. SVSM is a VSM-Based data tracking model integrated with IoT
and production flow in manufacturing processes. Similarly, no previous with the purpose of reducing the time spent searching for materials in
studies have been identified that tried to understand which metrics the process by workers (Phuong and Guidat, 2018). VSM-Based Logistics
would be the most relevant for better energy management and pro­ 4.0 integrate VSM and IoT to eliminate the gaps between the
duction processes or what their main interrelationships would be. current-sate and future-state maps of logistics activities (Boonso­
thonsatit et al., 2020).
Although all models found with integration of I4.0 technologies
2.1. VSM-based models found in literature presented innovations, none of them advanced with studies for the use of
such resources for better energy management, focusing on the moni­
Several models analysed in the study were found during the scoping toring and flow of the production process and location of materials, not
review, as shown in Table 1. Some of these only addressed integration being able to really promote better energy efficiency practices. Several
with I4.0 technologies and others only the application of energy, envi­ barriers to implementing these management models were detected in
ronmental and production process metrics. In none of the studied the scoping review, such as the need for technical knowledge for their
models was VSM extended to energy management and integrated with implementation, use and maintenance (Balaji et al., 2020), high in­
I4.0 technology to provide dynamic data collection and analysis in real- vestment costs, lack of qualified professionals (Pagliosa et al., 2021),
time, enabling the modeling of energy through time series in order to their application in real production conditions (Huang et al., 2019a),
allow better energy management, reduction of equipment failures, and cultural issues (Pasi et al., 2020), resource limitations (Ilangakoon et al.,
budget forecast related to energy purchase. 2021), legal, social and financial impacts (Mayr et al., 2018) and pro­
duction problems (Wagner et al., 2018).
2.1.1. VSM-based models integrated with I4.0 technologies
The models that addressed the integration of VSM with I4.0 tech­ 2.1.2. VSM-based models focused on energy efficiency
nologies presented as the integration of digital information in real-time The VSM-Based models that addressed the application of energy,
to VSM, being able to identify any waste in the productive operation environmental and production process metrics sought to address energy
(Benzi et al., 2018), in addition to improving the analysis and visuali­ issues with traditional VSM. The main objective was to improve energy
zation of processes (Huang et al., 2019b), thereby providing greater efficiency in companies by identifying the level of energy consumed
accuracy compared to traditional tools (Balaji et al., 2020), minimiza­ (Keskin et al., 2012) and to reduce this significantly (Muller et al., 2014)
tion of manufacturing lead times (Ramadan et al., 2020) and ability to by identifying the added and non-added energy values in the process
monitor the positions of objects within the factory (Tran et al., 2021). cycles (Müller et al., 2014) to improve management practices (Svensson
DVSM is a traditional VSM integrated with I4.0 technologies (Benzi and Paramonova, 2017). These models can be used both as a diagnostic
et al., 2018) to obtain diagnostic information in real-time (Balaji et al., and a tool to help with energy budgeting (Verma and Sharma, 2019) by
2020), capable of storing, analysing, visualizing (Huang et al., 2019a), identifying the largest sources of waste (Baysan et al., 2019).
and minimize the downtime of the processes (Ramadan et al., 2020), EVSM is a VSM-Based model integrated with energy metrics capable
focusing mainly on material tracking (Tran et al., 2021) and information of detecting non-value-added energy consumption (Keskin et al., 2012),
flow in real-time (Wagner et al., 2018). VSM4.0 is a holistic integration energy wastes (Muller et al., 2014), energy demands (Schönemann et al.,
2016), energy savings (Melsas and Rosin, 2017), energy consumption
Table 1 (Xie et al., 2018), overload and variability effects on energy consump­
VSM-Based models found in literature. tion (Baysan et al., 2019), value-added energy consumption, energy
Models Authors budgeting and economy savings (Verma and Sharma, 2019), allowing
the identification of the area with the greatest potential of savings
VSM and Industry 4.0
Dynamic VSM (DVSM) Tran et al. (2021); Abideen and Mohamad (2021);
through changes in operational behaviour (Cosgrove et al., 2018).
Balaji et al. (2020); Ramadan et al. (2020); Pasi et al. IEVA integrate EVSM with energy audits and energy balance charts,
(2020); Huang et al. (2019); Benzi et al. (2018); enabling a top-down approach (Bettoni et al., 2015), TVSM is a model
Wagner et al. (2018) that integrated VSM with Therblig (Sihag and Sangwan, 2019) to
VSM4.0 Pagliosa et al. (2021); Ilangakoon et al. (2021);
improve energy transparency and the reduction of energy wastes of the
Valamede and Akkari (2020); Fortuny-Santos et al.
(2020); Buer et al. (2018); Mayr et al. (2018); process (Jia et al., 2017), GVSM is integrated with Interpretive Struc­
Camgoz-Akdag et al. (2018) tural Modelling (IPS) and Fuzzy AHP to improve sustainable logistics
VSM-Based Logistics 4.0 Boonsothonsatit et al. (2020) business and social activities (Rukmayadi et al., 2016), VVLD is a model
Sustainable VSM (SVSM) Phuong and Guidat (2018) that integrated VSM, Material Flow Cost Accounting (MFCA), and Pinch
Analysis to measure the efficiency of use of production resources
VSM and Energy Efficiency
(Thanki and Thakkar, 2016), LCAVSM is eco-efficiency scheme based on
Energy VSM (EVSM) Baysan et al. (2019); Verma and Sharma (2019); integration of the Life Cycle Assessment (LCA) with VSM, focusing on
Cosgrove et al. (2018); Garza-Reyes et al. (2018); energy use in order to reduce environmental impacts (Estrada-Gonzalez
Melsas and Rosin (2017); Xie et al. (2018); Svensson et al., 2020).
and Paramonova (2017); Schönemann et al. (2016);
While the models presented are concerned with integrating energy
Muller et al. (2014); Müller et al. (2014); Keskin
et al. (2012) metrics into the VSM for better process energy management, they did
Integrated Energy Value Bettoni et al. (2015) not propose a model with satisfactory integration with I4.0 technologies
Analysis (IEVA) in order to meet all the current dynamics of the processes present in the
Therblig-embedded VSM Sihag and Sangwan (2019); Jia et al. (2017) productive processes, in view of the constant changes in production
(TVSM)
Green VSM (GVSM) Rukmayadi et al. (2016)
processes. Without this integration, new analyzes or even the fidelity of
Value - Value Load Thanki and Thakkar (2016) what was verified is limited to when the collections took place. Some
Diagram (VVLD) barriers can be highlighted, such as the lack of integration with I4.0
Life Cycle assessment Estrada-Gonzalez et al. (2020) technologies, which prevents real-time data collection, data analysis and
VSM (LCAVSM)
storage and access to information. Some of the information needed for

4
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

the studied models has to be calculated manually, which makes the the literature to be mapped to comprehensively and efficiently evaluate
procedure expensive, complex and laborious (Cosgrove et al., 2018), the existing literature on framework propositions (Nascimento et al.,
which may not reflect real values within dynamic processes (Xie et al., 2022), summarizing complex and heterogenous topics (Di Pasquale
2018). et al., 2020). The scoping review was separated into five steps (Arksey
and O’Malley, 2005): identifying the research questions, relevant
3. Research design studies, study selection, charting the data, and collating, summarizing,
and reporting. An extra qualitative step has been added, aiming to
The research methodology began with a scoping review (de Mattos improve the quality of analysis, called validating and data coding
Nascimento et al., 2022), a diagnosis of the problems found in inte­ (Danese et al., 2018). The scoping review has recently been applied to
grating the VSM tool with I4.0 technologies to make industrial processes several areas related to this study, e.g., VSM (Marin-Garcia et al., 2021),
more energy efficient that identifies the main energy, environmental I4.0 technologies (Wilhelm et al., 2021), manufacturing (Di Pasquale
and production process metrics, and the I4.0 technologies most et al., 2020), and others.
frequently integrated with LM models. This information was then sub­ For reasons of transparency, it is important to detail how the entire
mitted to specialists for focus group discussion. The focus group rounds review was carried out (Saunders et al., 2012). Also, identifying research
(Tong et al., 2007) aimed to analyse the importance of a lean model for questions (i), interdisciplinary discussions between authors, brain­
energy management and to determine which were the most important storming, and literature analysis were realized, aiming to understand
metrics and technologies to include in the model. the state-of-the-art of energy management models and energy efficiency
In the Delphi study stage, a questionnaire was prepared by the au­ issues in the industrial sector and determine the most used I4.0 tech­
thors to present the general objectives and most important metrics nologies and energy, environmental and production process metrics. A
selected by the corresponding focus group to another group of specialists reasonable review provided the theoretical basis (Saieg et al., 2018) and
(Okoli and Pawlowski, 2004). Using the Fuzzy Delphi technique, the identified research gaps (Webster and Watson, 2002). To identify rele­
results related to the opinions of the participants were analysed quan­ vant studies (ii), the research string was constructed (Saieg et al., 2018)
titatively (A Ishikawa, 1993) to identify and classify the most relevant with the aid of a word tree of relevant terms found in the literature
energy, environmental and production process metrics. A new ques­ (Gabriele et al., 2012), as shown in Fig. 3. A portfolio was built using
tionnaire was then prepared with the metrics identified by the Fuzzy documents listed in Scopus and Web of Science (WoS) electronic data­
Delphi analysis to determine the interrelationships between the energy bases, configured to execute the search by title, abstract and keywords.
metrics on environmental and production process metrics using the The first pair (VSM and I4.0) found 34 documents, the second pair
Fuzzy DEMATEL technique (Garcia-Buendia et al., 2022). The Fuzzy (VSM and Energy Efficiency) found 41 documents, and the search string
Delphi method combines Fuzzy set theory and the traditional Delphi using VSM, I4.0, and energy efficiency found 0 documents, generating
technique to improve the proficiency of expert judgement by reducing 75 documents before the application of inclusion and exclusion criteria.
uncertainties (A Ishikawa, 1993), while the Fuzzy DEMATEL method is Study selection (iii) was applied considering inclusion and exclusion
used to convert the cause and effect relationships (Garcia-Buendia et al., criteria, where to inclusion criteria, only English and peer-reviews
2022) by identifying the most important energy metrics criteria that journals and conference proceedings documents were retrieved as they
affect environmental and production process metrics criteria. are considered the most reliable sources for literature reviews (Saunders
The main advantages of using Fuzzy Delphi are related to saving time et al., 2012). All publications up to May 2022 were used in the study,
on the questionnaire, saving costs, reducing the total number of surveys, yielding a total of 40 documents. Any VSM and I4.0 pair papers without
and questionnaires increase the recovery rate, experts can fully express at least one I4.0 technology integrated with VSM, and any VSM and
their opinions, ensuring completeness and consistency of opinions and Energy Efficiency pair papers that did not demonstrate the application of
taking into account the ambiguity that cannot be avoided during the energy, environmental and production process metrics with VSM to
study (Hsi-Mei Hsu & Chen-Tung Chen, 1996). The advantages of using improve energy efficiency were discarded. Inclusion and exclusion
Fuzzy DEMATEL can be described as overcoming the inevitable uncer­ criteria were applied to reduce any researcher interference or bias
tainty, reducing the number of surveys, the semantic structure of fore­ during selection (Siddaway, 2014).
cast items can be explained, and the individual attributes of the expert The selected documents underwent a three-stage evaluation. Firstly,
can be described (Akira Ishikawa et al., 1993). The research, when using the titles, abstracts and keywords were read. Next, the methodology,
the integration of Fuzzy Delphi and Fuzzy DEMATEL, provides greater results and/or findings, and conclusions were read. The final stage
robustness in the results, allowing to extract from the interviewees a comprised a full reading of the selected documents. A further seven
deep understanding of which criteria are most relevant, their ordering documents were added to the final portfolio by snowballing comple­
and also which are the strongest interrelationships, providing a broad mentary research, yielding a total of 34 publications to be used in this
understanding of the importance of the resulting metrics, encouraging a research. The scoping review identified the most used VSM-based
better use on the proposed model. All selected metrics were inserted in models focused on energy efficiency, VSM-based models integrated
the RTEFM model and applied at the LabFaber 4.0 with the purpose of with I4.0 technologies, energy, environmental and production process
obtaining a depth empirical investigation of the application within a real metrics, and I4.0 technologies most applied in conjunction with VSM.
environment (Quelhas et al., 2019). Charting data (iv) was performed by defining which methods and
Face validity requires the approval of nonresearchers regarding the information would be used for analysis, with descriptive (Núñez-Merino
validity of a study, being the arguably the most robust way to establish et al., 2020) and content analysis (Jose Arturo Garza-Reyes, 2015) being
face validity is the involvement of domain experts, also known as subject chosen for this study. Validating and data coding (v) enabled the com­
matter experts, before (a priori), during, after (a posteriori), or parison of the collected informations, in addition to the discussions to
throughout the research (Lucko and Rojas, 2010). This research was solve possible differences, guaranteeing the reliability, minimizing er­
followed by a Focus Group (Yu et al., 2006), Delphi Study (del Caño and rors and bias of the evaluators (Caiado et al., 2022). Collating, sum­
de la Cruz, 2002), and a case study (Rojas and Dossick, 2008) to marizing, and reporting (vi) was performed using descriptive analysis of
establish face validity. An overview of the stages applied in the research the main keys of the research, along with content analysis of the
is provided in Fig. 2. VSM-based models and their implementation barriers.

3.1. Scoping review 3.2. Focus group

The scoping review executed in the first stage is a method that allows A focus group was used during the second stage. This is a research

5
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Fig. 2. Research design of the study.

6
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Fig. 3. Search strings.

technique that uses participant interaction to generate data on topics into the entire research project and facilitating discussions by making
previously defined by the researcher (D. L. Morgan, 1996) for discussion the participants more similar to each other (D. Morgan, 1997). Table 2
and to understand specialist perceptions (Krueger, 1994). This tech­ lists the experts selected to participate in the focus group.
nique can, therefore, be considered a qualitative method that closely Table 2 presents the seven specialists selected for their perceived
resembles the interview. Meetings must follow certain protocols with a knowledge due to their educational level, research and/or professional
moderator continually fomenting interaction among all the participants, experience (Nascimento et al., 2022) in engineering, technology or en­
who may or may not be influenced by the responses of the others ergy. Apart from these experts, all the sessions included a moderator to
involved (Oliveira and Freitas, 1998). lead and drive the discussion. There were four sessions with a total
The main advantages of using focus groups in research are ease of duration of 281 min. The first interaction discussed the current state of
execution, the ability to explore topics and generate hypotheses, the energy in the industrial sector and how the proposed model could help
opportunity to collect data from group interaction (Krueger, 1994), high to achieve better energy efficiency in industries. In the second and third
data speed, low cost compared to other methods, speed in providing the iterations, the discussion revolved around the I4.0 technologies that
results and allowing the researcher to increase the sample size whenever would be essential to apply in the model and how energy efficiency
necessary (Oliveira and Freitas, 1998). Focus group sessions should have could be achieved in an industrial environment. The fourth interaction
between three and twelve participants (Kontio et al., 2008) and their determined the energy, environmental and production process metrics
segmentation has the advantage of building comparative dimensions that would be the most relevant for better energy management and
simultaneously identified which dimensions related to each metric were
addressed.
Table 2
Participants specialised in Industrial Technologies possessed a vast
Focus Group experts.
knowledge and experience of over 21 years working on industrial pro­
Code Experience Specialities Role Country Expert jects related to applied mechanics, industrial automation, hardware and
A1 21 years Industrial Electrical Brazil Practitioner software. The Sustainable Management Systems specialist was an aca­
Technologies Engineer demic with 15 years of experience who worked as a researcher and
A2 15 years Sustainable Industrial Spain Academic university lecturer in the areas of BIM, Lean Systems, Circular Economy,
Management Engineer
Systems
Industry 4.0, Lean 4.0 and Construction Life Cycle Project Management.
A3 23 years Industrial Electrical Brazil Practitioner The Energy Management Systems specialist had 7 years of experience
Technologies Engineer working directly in the energy sector, in energy efficiency, electric
A4 20 years Information Computer USA Practitioner mobility and hydroelectric and wind power plant projects. The IT
Technologies Networker
specialist had 20 years of experience in software development for
A5 5 years Industrial Electrical Brazil Practitioner
Planning Engineer management systems. The Industrial Planning specialists had 5 years of
A6 7 years Energy Mechanical Brazil Practitioner experience working in project management and planning.
Management Engineer
Systems
A7 5 years Industrial Civil Brazil Practitioner
Planning Engineer

7
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

3.3. Questionnaire design and expert panel between the energy metrics on environmental and production process
metrics. It is recommended that the same questionnaire is submitted to
The third stage considered the design and testing of the question­ the experts more than once in each round for their predictions to
naire used in the research and consisted of different phases to classify converge (Akira Ishikawa et al., 1993). Two questionnaire submissions
and clarify all the study information for the participants (Garcia-Buendia per round were established for the present study, giving a total of four
et al., 2022). The first phase dealt with demographic information such as submissions to be answered by the same participants.
each specialist’s current role, total experience time and time performing
their current role. The second phase elicited responses from the partic­ 3.4.1. Fuzzy theory
ipants to evaluate the metrics according to the questions posed. Fuzzy set theory introduced the concept of membership function to
Before the discussions started, a brief initial explanation was given of handle different linguistic variables problems (B. Chang et al., 2011).
the proposed conceptual model and the evaluated metrics’ general Chang et al. (2000) state that this theory was adopted to convert highly
definitions and how they would be used. The entire content of the uncertain linguistic preferences into quantitative values that continued
questionnaires was previously tested among the authors to identify any to be based on human choices and maintained their qualitative char­
inconsistencies, with preliminary access to the questionnaires being acteristics. The defuzzification process reduces a series of conclusions of
carried out for reading, answering and sending to guarantee that all the variable pertinence to a single output point (Cheung et al., 2005).
information was unambiguous. Although there is no rules regarding the
minimum number of experts needed for a Delphi study (Moktadir et al., 3.4.2. Fuzzy Delphi
2020), the recommendation is for ten to eighteen participants’ opinions The Fuzzy Delphi method integrates Fuzzy set theory with the
to be collected to reach a consensus (Okoli and Pawlowski, 2004). traditional Delphi method to reduce any uncertainties related to the
Twenty-four specialists who had not participated in the focus groups specialists’ preferences, thereby improving the quality of the results
were selected using established criteria: a minimum of 12 years of obtained in the research (A Ishikawa, 1993). It was developed by Helmer
experience in the industrial or academic sector, with specialities related and his associates as a long-term forecasting method that required
to management, engineering, or energy. Table 3 gives the characteristics repeated surveys of experts for the forecast values to converge (Akira
of the panel of experts who participated in the study. Ishikawa et al., 1993).
Table 3 lists all the selected specialists who participated in a two- Table 4 gives the triangular Fuzzy numbers that correspond to the
round survey. Of these, 79.17% had over 15 years of professional linguistic terms. Tsai et al. (2020) demonstrated an assumption that the
experience and 83.33% had an academic background in engineering, significance value of an attribute b is rated by a respondent a as j = (xab ,
which improved the quality of the responses. In the first round, a 5-point yab ,zab ), a = 1, 2, 3, …, n and b = 1,2,3,…,m. Thus, the jb weight of the b
Likert scale from 1 = irrelevant to 5 = extremely relevant was used to attribute is computed as jb = (xb , yb , zb ), where xb = min (xab ),
select the most relevant metrics for inclusion in the model. In the second ∏
yb = ( n1 yab )
1/n
and zb = max (zab ). According to Wu et al. (2016), the
round, respondents had to indicate the influence of energy metrics on value of the convex combination Db is generated using α cut, and in
environmental and production process metrics using the linguistic scale common situations, as shown by Lee et al. (2018), the adopted α value is
NI, LI, MI, HI and EI (‘no influence’, ‘low influence’, ‘moderate influ­ normally 0.5 but can be adjusted depending on the experts’ level of
ence’, ‘high influence’ and ‘extreme influence’, respectively). All optimism or pessimism defined as 0 or 1. The calculation of the value of
twenty-four experts participated in both rounds. the convex combination can be expressed by Equation (1):
∫
3.4. Fuzzy Delphi and fuzzy DEMATEL study Db = (ub , lb ) = [λub + (1 − λ)lb ] (1)

In the fourth stage, the Fuzzy Delphi and Fuzzy DEMATEL techniques where:
were used to analyse the answers to the questionnaires. Two distinct
ub = zb − α(zb − yb )
rounds were carried out, the first to understand which energy, envi­
ronmental and production process metrics were the most relevant ac­
lb = xb − α(yb − xb )
cording to the experts’ perceptions. After discarding the less relevant
metrics, the second round identified the strongest interrelationships b = 1, 2, 3, …, m

Table 3 α = Expert’s level of optimism

Composition of the expert panels. λ = Decision-maker’s level of optimism
Lambda (λ), besides expresses a decision-maker’s level of optimism,
Characteristics n (%)
also stabilises the expert group’s radical judgements. Therefore, δ =
Experience ∑n Db
From 12 to 15 years 5 (20.83) a=1 n is the filter threshold of required attributes. If Db ≥ δ, attribute b
From 15 to 25 years 10 is accepted, otherwise it must be rejected (Garcia-Buendia et al., 2022).
(41.67)
Over 25 years 9 (37.50)
3.4.3. Fuzzy DEMATEL
Role
Administration (Director, Manager and Coordinator) 13 DEMATEL is an extended method for building and analysing struc­
(54.17) tural models that can explain the influence relationships between
Technical (Application Engineer, Industrial Engineer and Technical 8 (33.33)
Consultant)
Commercial (Sales Engineer) 2 (8.33)
Research 1 (4.17) Table 4
Expert Corresponding triangular Fuzzy numbers for relevance scale.
Practitioners 22
Scale Linguistic term Triangular Fuzzy Numbers (TFN)
(91.67)
Academics 2 (8.33) 1 No relevance (NR) (0; 0; 0.25)
Qualification 2 Low relevance (LR) (0; 0.25; 0.50)
Engineering 20 3 Moderate relevance (MR) (0.25; 0.50; 0.75)
(83.33) 4 High relevance (HR) (0.50; 0.75; 1.0)
Others 4 (16.67) 5 Extreme relevance (ER) (0.75; 1.0; 1.0)

8
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

complex criteria (B. Chang et al., 2011). This method is useful for rvkij = Right normalised value computed
visualizing complex causal relationship structures with matrices or di­ Step 4. Integrate the crisp values (̃
x):
graphs and is able to convert the relationship between the criteria causes
and effects into a comprehensible structural model of the system x1ij + xij2 + … + xkij
xij =
̃ (6)
(Falatoonitoosi et al., 2013). K
Table 5 gives the linguistic terms and their corresponding triangular
Fuzzy numbers. Fuzzy DEMATEL uses the defuzzification technique to where:
transform qualitative information into Fuzzy linguistic data, with the
process generating crisp values from Fuzzy numbers. Then the left and xkij = Normalised crisp value computed
right values are computed by minimum and maximum fuzzy numbers. K = Total number of respondents
The fuzzy membership functions fijk = (flkij , fmk ij , frkij ) are used to generate Step 5. Arrange the generalised direct relation matrix (G):
[ ]
the total weighted values. Assuming that k experts participated in the G= ̃xij I×J (7)
evaluation process, the fijk signifies the fuzzy weight of the ith attribute’s
effect on the jth attribute as assessed by expert kth . The crisp values ate where:
then presented in a total direct matrix, allowing a diagram to be drawn
to simplify the analytical results. The cause and effect groups containing xij = Integrated crisp values
̃
certain attributes represent the structured interrelashionships and Step 6. Compute the normalised total direct relation matrix (T):
important effects. A set of attributes is proposed (F), and certain pairwise T =τ ⊗G (8)
interrelashionships are used to create the mathematical relations (Tsai
et al., 2020). The proposed Fuzzy DEMATEL method is divided into 6 1
steps, as shown below: τ= ( I ) (9)
∑
max xij
̃
i=1
Step 1. Normalise the Fuzzy numbers:
⎡ ( ) ( ) ( )⎤ where:
( ) lkij − min lkij mkij − min mkij rijk − min rijk
k k k k
Fij = flij , fmij , frij = ⎣ , , ⎦ (2)
Δ Δ Δ G = Generalised direct relation matrix
xij = Integrated crisp values
̃
( ) ( )
Δ = max rijk − min lkij , mkij , rijk (3)
3.5. Case study
where:
The conceptual model was elaborated using the most relevant met­
rics resulting from the Fuzzy Delphi and Fuzzy DEMATEL study,
lkij = Left TFN values
together with the I4.0 technologies discussed by the focus group. This
mkij = Mean TFN values model was submitted to a case study for empirical validation. A case
rijk = Right TFN values study is research approach that is used to generate in-depth, multi-
Step 2. Compute right (rv) and left (lv) normalised values: faceted understanding of a complex issue in its real-life context (Crowe
[ ] et al., 2011), especially when the limits between the phenomenon and
( ) fmk ij frkij the context are not clearly defined (Yin, 2009). Case study can be
k k
lvij , rvij = , (4)
1 + fmk ij − flkij 1 + frkij − fmk ij characterized in three main types: intrinsic, instrumemental, and col­
lective (Stake, 1995). An intrinsic case study is normally used to learn
where: about a unique phenomenon. The instrumental case study uses a
particular case to gain a broader appreciation of a phenomenon, and the
flkij = Minimum normalised Fuzzy number collective case study involves studying multiple cases simultaneously or
sequentially aiming to generate a broad analysis of a particular issue
fmk ij = Mean normalised Fuzzy number
(Crowe et al., 2011).
frkij = Maximum normalised Fuzzy number Case study can be used to explain, describe or explore events or
Step 3. Compute the normalised crisp values (x): phenomenon in the everyday contexts in which they occur, and can be
[ ( ) ( )2 ] classified in three typed regarding the research objective: descriptive,
lvkij 1 − lvkij + rvkij exploratory, and explanatory (Yin, 2009). Descriptive case study
xkij = ( ) (5) describe the phenomenon with its context, exploratory deals with
1 − lvkij + rvkij little-know issues, trying to define hypotheses or propositions for future
research, and explanatory aims explain relations of cause and effect from
where: theory (Quelhas et al., 2019). A single, intrinsic, and explanatory case
study was applied with the purpose of verifying and validating the
lvkij = Left normalised value computed functioning and behavior of the RTEFM model in a real manufacturing
plant at the LabFaber 4.0 through practical tests. The implementation of
the case study followed the following steps, as shown in Fig. 4.
Table 5 After the severe and careful selection of the most relevant metrics
Corresponding triangular Fuzzy numbers for influence scale. and application in the conceptual model, the research sought, using the
Scale Linguistic terms Triangular Fuzzy Numbers (TFN) case study, to apply this model at the LabFaber 4.0, following the steps
NI No influence (0; 0.1; 0.3) shown in Fig. 4. The presentation of the model was premised on indicate
LI Low influence (0.1; 0.3; 0.5) to those responsible how the implementation and operation of the sys­
MI Moderate influence (0.3; 0.5; 0.7) tem would take place, seeking to understand the real and local needs in
HI High influence (0.5; 0.7; 0.9) relation to energy issues, subsequently analyzing all the costs involved
EI Extreme influence (0.7; 0.9; 1.0)
in the implementation and installation of the hardware responsible for

9
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Fig. 4. Overview of RTEFM implementation steps.

the measurements, as well as the times related to the necessary stops of determined by the user; Energy Usage is the amount of energy used by
machines and equipment. equipment; Energy Cost is the cost of energy used by a machine or
The identification of which machines were chosen for measurements during the industrial production process over a period of time deter­
and tests was based on the productivity criterion, choosing those that mined by the user; Energy Input is the amount of energy consumed at the
would possibly have the highest energy consumption. An analysis of the process input, excluding losses; Energy Value is the amount of energy
impacts to adapt the existing infrastructure for correct and satisfactory used per unit of product or batch; Energy Utilisation is the percentage of
communication was carried out, followed by the installation of IoT the process’s total energy consumption used in the execution of a given
measurement sensors, integrating them into the proposed management activity; Non-Value-Added Energy is the amount of energy consumed by
system. Training and workshops were carried out with the aim of pre­ activities that do not generate value; Value-Added Energy is the amount
paring users to operate and monitor the system’s dashboard, with all of energy consumed by activities that generate value; Energy Cost per
parameterization and selection of desired metrics being released and Item is the energy cost per unit of product or batch.
allowed, that is, as long as the parameters exist, the user would be free to Environmental and production process metrics are defined as fol­
use energy, production process, and enviromental metrics, and also lows: Cycle Time is the amount of time required to produce a given item
create new metrics according to their needs. Finally, the monitoring and or batch; Lead Time is the total amount of time taken to deliver a product
analysis of information would be available on the system panel through along the value chain; Greenhouse Gas Emissions is the amount of CO,
a web platform. CO2, H2S, NH3, NO2, SO2 and dust emitted during the production
process; Power Required is the electrical power (apparent, active and
4. Analysis and results reactive) required by the machine to perform activities; Waiting Time is
the amount of time spent between productive stages, i.e., downtime, and
This section reports the qualitative and quantitative results obtained Water Consumption is the amount of water used during the production
in the study, presents and discusses the main findings and proposes a process.
new conceptual model called RTEFM, which has the ability to collect, All the I4.0 technologies and energy, environmental and production
store and visually display the energy flow map of machines and indus­ process metrics were extensively discussed by the focus group, as was
trial production process equipment in real time. the need for a new conceptual model of energy management for in­
dustries. The purpose was to understand the current scenario of energy
efficiency in the industrial sector and to identify the metrics that would
4.1. Metrics, I4.0 technologies and opportunities in energy management
represent the best combination for energy management according to
expert perception. The experts expressed major concerns about energy
The scoping review included an analysis of the final portfolio of 34
losses during energy transformation processes and, in the industrial
documents, all of which were read in full and, as can be observed in
environment, the increasing complexity surrounding obtaining infor­
Appendix A, the main energy, environmental and production process
mation on the energy consumption of installed equipment, which would
metrics, and main I4.0 technologies found in the document models were
limit stakeholders’ ability to evaluate the flow of energy in processes and
selected for discussion by the focus group.
hamper their decision-making.
Energy metrics are defined as follows: Energy Consumption is the
Therefore, the greatest advantage of our conceptual model would be
energy consumption of machines or processes during a period of time

10
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

its ability to inform the operator or any other interested party about the B, in which the linguistic scales NR, LR, MR, HR and ER (‘no relevance’,
amount of energy being consumed by any particular machine or ‘low relevance’, ‘moderate relevance’, ‘high relevance’ and ‘extreme
equipment in real time, which would allow to obtain both an overview relevance’) were used, as described in Table 4. The values 0.5 and 1
and a snapshot of the entire energy flow in the process. For experts, one were adopted for α and λ respectively (P.-T. Chang et al., 2000). The
of the main barriers at the current time would be the difficulty of col­ calculated threshold (δ) value was 0.811 and the metrics whose convex
lecting data on the various extant industrial environments; however, combination (Db) showed values lower than δ were rejected, as can be
they agree that these barriers would only be temporary thanks to de­ observed in Table 6.
velopments in I4.0 technologies. As Table 6 shows, only three of the sixteen metrics analysed were
The focus group recognized that implementing Industry 4.0 tech­ rejected, namely Lead Time, Non-Value-Added Energy and Value-Added
nologies could potentially enhance an organization’s market competi­ Energy. The most relevant metric for the technical dimension was En­
tiveness, despite the significant investments required to adapt processes. ergy Consumption; for the financial dimension, Energy Cost; for the
According to the experts, the I4.0 technologies in the literature with the governance dimension, Energy Consumption per Defective Item and for
greatest synergies for inclusion in the proposed conceptual model are: (i) the environmental dimension, Water Consumption.
IoT for sending and receiving data from machines and industrial pro­
duction processes; (ii) Edge Computing to reduce response times during
data storage and processing; (iii) Cloud Computing for storing records 4.3. Interrelationships between energy metrics on environmental and
and information on machines and equipment; (iv) Big-Data Analytics production process metrics
(BDA) to process and analyse the data obtained; (v) Web Technologies as
a digital interface between the user and process data; (vi) Cyber-Security As Table 6 indicates, after identifying, classifying and selecting the
to ensure the security of all information transmitted over the network; most relevant energy, environmental and production process metrics, to
(vii) Interoperability to promote complete communication between all answer RQ3, a new study was conducted to determine the strongest
devices on the network using protocol and format neutral information interrelationships between the studied metrics. The Fuzzy DEMATEL
structures. technique was applied in this phase to evaluate the influence between
The focus group analysed the energy, environmental and production the energy metrics on environmental and production process metrics.
process metrics found in the literature and discussed those that would be The first step in a Fuzzy DEMATEL analysis is to shape the initial
the most efficient and relevant for application in an energy management direct relation matrices I using the information collected from the spe­
model and their corresponding dimensions. The specialists had the in­ cialists. Appendix C gives an example using the data collected from
clusion of the Energy Consumption per Defective Item metric in the Respondent 1. The responses from the respondents are used to generate
model, defined as the energy consumed per defective or damaged part individual initial direct relation matrices I. As can be seen in Table 5, as
and this was presented for analysis along with the other metrics. Only previously mentioned, the experts were asked to classify the influence of
metrics that all focus group participants considered unnecessary were the energy metrics on environmental and production process metrics on
discarded. The metrics selected as the most relevant by the focus group the linguistic scale NI, LI, MI, HI, EI (‘no influence’, ‘low influence’,
can be seen in Fig. 5 with their dimensions. ‘moderate influence’, ‘high influence’ and ‘extreme influence’).
In this context, according to the information obtained from the The design of the Fuzzy linguistic variables converts the initial direct
scoping review and the extensive focus group discussion, the answer to relation matrices I into triangular Fuzzy numbers using the data in
RQ1 is that the use of RTEFM would enable companies to better manage Table 5. Each linguistic variable is replaced by its corresponding num­
their energy flow and also model energy consumption in their processes, ber, as shown in Appendix D. The defuzzification process is conducted
as an effective energy management model is a major driver of cost using Equations (2) and (4). Equation (5) uses the resulting crisp values
reduction, increased performance and improved process sustainability to generate the direct relation matrices D in Appendix D.
indicators. Equation (6) is used to calculate the average values of the direct
relation matrices D. Appendix E presents the generalised direct relation
matrix G for energy, environmental and production process metrics,
4.2. Most relevant energy, environmental and production process metrics built with Equation (8). The total direct relation matrix T obtained by
applying Equations (8) and (9) is presented in Table 7, which shows the
To answer RQ2, the Fuzzy Delphi technique was applied to identify interrelationships between the evaluated metrics.
the most relevant energy, environmental and production process metrics Table 7 shows the strength of the relationships between the energy
according to Equation (1). The experts’ responses are given in Appendix metrics on environmental and the production process metrics based on

Fig. 5. Metrics and dimensions.

11
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Table 6
Fuzzy Delphi method results after screening.
Dimension Metrics Code ub lb Db Decision

Technical Energy Consumption E1 0.938 0.312 0.938 Approved

Energy Usage E2 0.916 − 0.041 0.916 Approved
Energy Input E4 0.883 − 0.383 0.883 Approved
Energy Value E5 0.866 − 0.366 0.866 Approved
Energy Utilisation E6 0.858 0.017 0.858 Approved
Cycle Time P1 0.919 − 0.044 0.919 Approved
Lead Time P2 0.500 0.000 0.500 Rejected
Power Required P4 0.877 − 0.377 0.877 Approved
Financial Energy Cost E3 0.921 − 0.046 0.921 Approved
Energy Cost per Item E9 0.899 − 0.024 0.899 Approved
Waiting Time P5 0.832 − 0.332 0.832 Approved
Governance Non-Value-Added Energy E7 0.500 0.000 0.500 Rejected
Value-Added Energy E8 0.500 0.000 0.500 Rejected
Energy Consumption (Defective Item) E10 0.842 − 0.342 0.842 Approved
Environmental Greenhouse Gas Emissions P3 0.850 − 0.350 0.850 Approved
Water Consumption P6 0.869 − 0.369 0.869 Approved
Threshold δ 0.811

information storage have layers of security via Cyber-Security. On the

Table 7
Web platform, the user can select metrics by dimension, relevance, in­
Total direct relation matrix T.
fluence or individually. New metrics can be registered providing that the
P1 P3 P4 P5 P6 parameters available during collection are respected. The entire archi­
E1 0.207 0.221 0.191 0.154 0.168 tecture is shown in Fig. 6.
E2 0.186 0.188 0.168 0.145 0.141 As shown in Fig. 6, the great benefit of the model is its flexibility to
E3 0.142 0.163 0.156 0.133 0.131
adapt to different customer needs since its data collection structure via
E4 0.101 0.188 0.167 0.111 0.093
E5 0.142 0.187 0.154 0.171 0.151 IoT sensors has a data structure protocol and neutral format. Interop­
E6 0.209 0.224 0.161 0.238 0.169 erability between systems makes communication simpler. Another
E9 0.182 0.183 0.171 0.200 0.149 important point regarding the proposed model is its ability to allow the
E10 0.124 0.138 0.124 0.187 0.147 decision-maker to visualise the industrial process’ entire energy flow
map (see Fig. 7). Users can use the stored energy consumption data to
expert opinions. Higher values indicate a stronger interrelationship. As model equipment energy use with time series and, thus, detect anoma­
suggested by Garcia-Buendia et al. (2022), a percentage analysis was lies and prevent any possible process failures before they occur.
implemented to facilitate the interpretation and understanding of the As the example in Fig. 7 shows, users have an overview of the energy
results, with the percentiles 80 and 60 computed as limits representing flow in the selected process and can configure the information that they
strong and moderate influence on the relationship (0.188 and 0.170, want to select, verify and analyse. If necessary, they can also determine
respectively). See Table 8. the individual consumption of a particular machine or piece of equip­
Thus, metric relationships with values above 0.170 (percentile 60) ment. A complete visualization of the energy flow in processes allows for
were considered to have a moderate influence (light grey) and metric more rapid and precise interventions and also helps to improve energy
relationships with values above 0.188 (percentile 80) were considered efficiency in projects and act on the exact points where value is being
to have a strong influence (dark grey). Any interrelationships below lost. Another advantage is that future energy consumption can be fore­
0.170 was considered to have an insufficient influence. cast, so the economics of contracting the energy required can be
improved.
4.4. Proposal for real-time energy flow mapping
4.5. Testing and operation in a real-world environment
The proposed conceptual model is based on real-time data collection
using IoT, information storage in Edge Computing and Cloud A single case study was carried out at the LabFaber 4.0 to carry out
Computing, data analysis using BDA and presentation of information via an alpha test of the metrics selected in the research, together with tests
a Web platform that can be customised to user needs. All traffic and related to customization according to the user’s input parameters,

Table 8
Relevant relationships between energy metrics on environmental and production process
metrics.

12
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

is operating at full load. The historical values of consumed energy can be

observed in the test performed, and the user can select the period he
wants to analyse the graphs.
This dashboard is applied to any metrics belonging to the model.
Through this analysis, the user or stakeholders can check online on the
RTEFM, thanks to the Cloud Computing and Web platform, whether the
machines and equipment are operating as expected, comparing both
with databook informations as well as its historical series. The calcula­
tion of the cost of energy is done by multiplying the energy consumed
and the price of the contracted energy, which in this case was R$0.75 per
1kWh. The entire dashboard can be configured by the user, allowing in
addition to the selection and creation of new metrics, also the
arrangement of the entire layout for better visualization. At the bottom
of the dashboard, a VSM of the process is shown, with the purpose of
facilitating the user in selecting which steps he wants to carry out a more
specific monitoring, or else, if he wants, the entire system.
In this way, the RTEFM model was applied in a real test of a conveyor
belt, which is part of a manufacturing process. The results obtained were
consistent with expectations, as well as its ability to promote a complete
visualization of the system. Stakeholders, using the tool, will be able to
understand possible energy and production process improvements, fault
corrections, budget forecasting, and others. In a simple operational
demonstration, the test was able to allow the visualization of the energy
consumed (technical dimension), energy cost (financial dimension), and
greenhouse gas emissions (environmental dimension).
Fig. 6. RTEFM, technological model.
5. Discussion
functionality and ergometry. The system, based on a web platform, was
integrated with IoT sensors, as shown in Fig. 6, that collected real-time The present study was conducted using qualitative and quantitative
data regarding line voltages, phase voltages, phase current, frequency, techniques to understand the state-of-the-art of LM models focused on
active power, reactive power, apparent power, power factor, energy energy and how these models could improve energy efficiency in in­
active, reactive energy, apparent energy, and harmonic distortions. dustries (RQ1); to identify the I4.0 technologies whose inclusion in the
Although the RTEFM can monitor the entire process, from the acquisi­ model is essential and determine the most relevant energy, environ­
tion stage to product delivery, a single case study was applied to the mental and production process metrics (RQ2) and their interrelation­
conveyor belt of the input stage to apply and verify the model. This ships (RQ3).
conveyor belt is driven by a 4-pole squirrel cage motor powered by a RTEFM fills the gap in the literature regarding VSM-based LM models
220Vac three-phase system with a frequency of 60Hz. The nameplate to improve energy efficiency by monitoring energy consumption (Xie
data of the applied motor informed by the manufacturer are electric et al., 2018) through the dynamic energy flow mapping of production
power of 1.5 kW (i); nominal rated current 6.65A (ii); power factor of processes (Sihag and Sangwan, 2019). The use of I4.0 technologies such
0.71 (iii); and electrical efficiency at 0.86% (iv). The test was carried out as IoT (Buer et al., 2018), BDA (Valamede and Akkari, 2020) and Cloud
within an operational period of 4 h, starting at around 8:00 a.m. and Computing (Ilangakoon et al., 2021) enables the proposed model to
ending at noon. A whole load was placed on the machine to test it as meet the needs of lean tools to leverage energy practices with real-time
close to operational reality as possible. Energy consumption, energy data traffic.
cost, greenhouse gas emissions, cycle time, and waiting time metrics The proposed model can identify the area with the greatest potential
were selected for testing, as shown in Fig. 8. for energy savings through changes in operational behaviour (Cosgrove
To calculate the active power of electric motors in three-phase sys­ et al., 2018) that reduce total energy consumption and environmental
tems, Equation (10) is applied: impacts (Estrada-Gonzalez et al., 2020). The model is also capable of
detecting non-value-added energy consumption (Keskin et al., 2012),
P3∅ = √3 × VL × In × cos φ (10) identifying energy waste (Jia et al., 2017), and analyse the effects of
overloads and energy variability (Baysan et al., 2019), helping
where: decision-makers to improve energy savings and budget forecasts (Verma
and Sharma, 2019). As one of the functions of the model is energy
P3∅ = Active power in three-phase systems modelling, a history of the machine’s energy consumption behaviour
VL = Line voltage will be present in the database, making it possible to identify any
In = Rated current behavioural changes in the energy consumption of the process, that is, if
cos φ = Power factor the machine has not undergone changes related to load or process speed,
an increase in energy consumption can indicate maintenance needs,
The power informed by the manufacturer is the rotational mechan­ which, if not detected, can cause breakdowns and failures, in addition to
ical power provided by the motor, that is, it is the power after the unnecessary energy consumption, directly impacting the environment.
transformation of electrical energy into mechanical energy, therefore, However, the model is not only limited to these measurements; it is
taking efficiency into account. Thus, when applying Equation (10), an possible and feasible to add metrics related to greenhouse gas emissions
active power of 1.799 kW and an apparent power of 2.53kVA are ob­ or water consumption in the processes, generating its time series,
tained. The calculation of the consumed energy is performed by multi­ allowing a complete analysis and monitoring of the entire production
plying the apparent or active power with time, usually expressed in kWh process, impacting directly and positively on environmental issues.
unit. Comparing with the values obtained by the model, the energy All the information generated by the model is available dynamically
consumed is very close to the motor nominal, indicating that the system due to the use of I4.0 technologies to improve real-time data collection

13
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Fig. 7. Overview of the proposed VSM-based model.

(Buer et al., 2018). IoT sensors collect all the available data from ma­ energy efficiency brings both technical and financial benefits (Verma
chines and equipment (Camgoz-Akdag et al., 2018) with the information and Sharma, 2019). The Fuzzy Delphi study identified thirteen energy,
then sent to Edge Computing to reduce network latency and congestion environmental and production process metrics whose technical and
generated by their offload from end-device tasks (Fernández-Cerero financial dimensions afforded the metrics the highest convex combina­
et al., 2020). Cloud Computing is used to store received data packages tion values: 53.85% and 23.08% participation, respectively. Therefore,
(Wagner et al., 2018) for further data and information analysis using companies using the proposed model are expected to be able to improve
BDA (Mayr et al., 2018). their production processes’ technical and financial characteristics as
RTEFM and the energy, environmental and production process spiralling energy demand is directly influenced by economic growth
metrics are available via Web Technologies for data integration (Jir­ (Keskin et al., 2012).
kovsky et al., 2017). Cyber-Security technology is implemented to pro­ The three most relevant energy metrics identified in the Fuzzy Delphi
tect networks and data and guarantee the reliability of communications, study were Energy Consumption, Energy Cost and Energy Usage.
resources and information management (Valamede and Akkari, 2020). Together, these metrics represent approximately 41.76% of the most
Interoperability between systems is essential as it enables the model to cited energy metrics found in the scoping review, followed by Energy
be modular and flexible, and implementable in any type of industrial Consumption at only 17.58%. Cycle Time, Power Required and Water
equipment (Hermann et al., 2015). Consumption were identified as the three most relevant environmental
The results obtained from the scoping review and the focus group and production process metrics. Cycle Time was the most cited metric in
provided a solid basis for understanding the barriers to implementing a the scoping review with approximately 27.27%. It is interesting to note
lean energy management model, including the identification of the most that, although relevant according to the analyses, Power Required
used energy, environmental and production process metrics. The most (Bettoni et al., 2015) and Water Consumption (Rukmayadi et al., 2016)
important barriers to implementing the model were found to be the lack were only mentioned once each in the literature.
of organisational procedures, the lack of adequate knowledge and the Fuzzy DEMATEL identified sixteen interrelationships between en­
company’s limited financial resources (Palm and Thollander, 2010), and ergy metrics on environmental and production process metrics, on the
these were subsequently categorised as organisational, behavioural and influence scale adopted, in which percentiles 80 and 60 were computed
economic dimensions (Johansson and Thollander, 2018). as limits representing strong and moderate influence, respectively
The scoping review showed that using VSM-based models to boost (Garcia-Buendia et al., 2022). It should be highlighted that the Energy

14
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Fig. 8. RTEFM dashboard.

Utilisation metric appears in both first and second place in Table 7 due to study to assess the need to monitor the relationship between energy and
its powerful influence on Waiting Time and Greenhouse Gas Emissions. environmental metrics as emissions due to energy demand have risen in
The relationship between Energy Consumption and Greenhouse Gas recent years (Schönemann et al., 2016), with industry responsible for
Emissions ranked third. Fig. 9 presents a summary of how and which 23% of CO2 emissions in the world (IEA, 2021a). Another important
metrics were selected. interrelationship demonstrates that attention should be paid to energy
Fig. 9 gives a concise overview of how all the metrics were analysed when it is not actually producing to identify losses that may be pre­
and selected in the scoping review. Energy Consumption per Defective vented or reduced (Bettoni et al., 2015). Table 9 shows a comparison
Item was added at the suggestion of the focus group. Although Energy between of the RTEFM and the others VSM-Based models with a focus on
Cost was the most relevant financial dimension and Water Consumption energy efficiency found in literature.
the most relevant environmental dimension, according to Table 8, none To summarise, as shown in Table 9, the RTEFM was developed using
of these metrics had a strong or moderate influence on any other metrics various data sources and expert’s perceptions. Both qualitative and
in the study. Regarding the financial dimension, the respondents quantitative methods were applied for greater rigour in the analyses to
considered that Energy Cost per Item would have a stronger interrela­ identify all the I4.0 technologies, energy, environmental and production
tionship with most environmental and production process metrics. process metrics, used in the previous research through a scoping review
The environmental dimension Greenhouse Gas Emissions metric (de Mattos Nascimento et al., 2022) and a focus group (Machado et al.,
presented the highest number of interrelationships with energy metrics, 2021). Fuzzy Delphi was then used to identify the most relevant metrics
demonstrating the concern of the specialists who participated in the (Garcia-Buendia et al., 2022) and Fuzzy DEMATEL to establish the

15
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Fig. 9. Metrics selection, overview.

Table 9
Comparison between VSM-Based models.
Models I4.0 technologies Energy approach Production Process Environmental
approach approach approach
Consumption Financial Modelling and Customization
performance

RTEFM X X X X X X X
EVSM X X
IEVA X X
TVSM X X X
GVSM X X X
VVLD X X X X
LCAVSM X X X

strongest interrelationships (Ashtiani Araghi and Vosoughifar, 2023), through energy control environmental and social impacts, and others.
being applied in a case study at the LabFaber 4.0 for a face validity Therefore, the model would benefit not only the organization but society
(Rojas and Dossick, 2008). as a whole.
Managers, technicians and users, or all interested stakeholders, will
be able to monitor the entire flow of processes, from metrics of pro­ 6. Conclusions
ductive processes as well as metrics related to energy, and can even
create their metrics from existing parameters to allow complete and The reserach managed to go from the literature review to the
quick decision-making. This monitoring will be carried out in real-time application of a new LM model in a satisfactory way, being able to
and will also allow, through energy modelling, to compare with the integrate energy, environmental and productive processes metrics in
historical behaviour of the machines from time series. The model will favor of energy efficiency in the processes. This entire model was able to
help organizations in an unprecedented way in all aspects of the ESG obtain information in real time with web access to data thanks to the
(Environmental, Social and Governance), as it will allow better use of integration with I4.0 technologies. The model was tested by users of the
the employees’ work, improvement and technical training, better con­ company LabFaber 4.0, where the results were compared with the
trol of processes, reduction losses, reduction of machine downtime, equipment databook, satisfactorily promoting a detailed dashboard
greater capacity for budget forecasting for energy acquisition, reduction referring to energy consumption, energy cost, greenhouse gas emissions,

16
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

cycle time and waiting time. For this, a RTEFM model has been pre­ addition to increasing productivity, the development and adoption of
sented that integrates IoT, Edge Computing, Cloud Computing, BDA, cleaner production technologies by stakeholders, promoting the effi­
Web Technologies, Interoperability and Cyber-Security technologies, cient use of natural resources in processes productive processes,
and the most relevant energy, environmental and production process including reducing the emission of greenhouse gases, causing a sys­
metrics. The integration of a VSM-based approach with I4.0 technolo­ tematic reduction of environmental impacts.
gies and energy efficiency metrics was explored to answer the study’s One limitation of this study is the concentration of respondents in a
three RQs. single country, Brazil. Therefore, the responses to the questionnaires
After selecting the metrics through the scoping review, the study may have shown some regional bias in the experts’ perceptions of the
included four sessions with specialists in focus groups that discussed the metrics’ relevance and influence, and it is interesting to carry out further
current state of energy in the industrial sector, the proposed manage­ research with experts from other countries. However, this is a minor
ment model, how to achieve energy efficiency in industries, I4.0 tech­ limitation as the criterion for their selection was that they should be
nologies and energy, environmental and production process metrics. qualified professional specialists with long experience in the industry.
The metrics selected after extensive discussions by the focus group Another limitation is that the model was applied to a conveyor belt that
participants were used in two rounds of questionnaires for a subsequent is part of a much larger process, thus isolating the results in a point
Delphi study to understand the respondents’ opinions and determine the analysis. Therefore, regarding future research, the RTEFM can be
most relevant metrics and their interrelationships. The resulting metrics applied in a complete process, allowing a complete analysis of the flow
were applied in the proposed model in conjunction with the I4.0 tech­ of information, materials and energy. One of the great advantages of the
nologies. A case study was applied to test and validate the functionality model is, precisely, its flexibility of implementation and integration,
of the model. since the entire system can be customised to the organisation’s needs
From a theoretical point of view, our investigation using a scoping while overcoming the concern about systems interoperability through
review revealed a lack of VSM-based lean energy management models protocol and format neutral information structures.
integrated with I4.0 technologies. Deep studies were carried out to
ascertain the state-of-the-art of energy efficiency in the industrial sector CRediT authorship contribution statement
and to identify the main lean models focused on energy management,
the most discussed I4.0 technologies and the most used metrics. The Leonardo Fontoura: Conceptualization, Methodology, Formal
concept of sustainable development is clearly on the agenda around the analysis. Daniel Luiz de Mattos Nascimento: Project administration,
world and that issues related to efficient electricity consumption in Supervision. Julio Vieira Neto: Validation. Noelia Garcia-Buendia:
production processes and the reduction of energy costs are some of the Writing – original draft, preparation. Jose Arturo Garza-Reyes: Visu­
industrial sector’s main concerns. From a practical point of view, man­ alization. Gilson Brito Alves Lima: Data curation. Guilherme Luz
agers using the RTEFM tool would be able to map the entire energy flow Tortorella: Resources. Rodrigo Goyannes Gusmão Caiado: Investi­
in their production processes in real-time and online, using a Web pla­ gation. Marcelo Jasmim Meiriño: Writing – review & editing.
taform, thus enabling more accurate and faster decision-making and cost
reductions and increasing their market competitiveness. Allied to this, Declaration of competing interest
the tool would also allow the modeling of energy consumption through
time series, making it possible to identify equipment failures due to The authors declare that they have no known competing financial
behavioral changes in its electrical consumption, thus reducing machine interests or personal relationships that could have appeared to influence
downtime. Another important advantage in using the model would be the work reported in this paper.
the identification of equipment and machines with less clean technol­
ogy, using continuous and preventive monitoring of production pro­ Data availability
cesses. Therefore, RTEFM, by constantly monitoring energy,
environmental and production process metrics, would provide, in No data was used for the research described in the article.

Appendix A. Energy metrics, environmental and production process metrics, and I4.0 technologies

References

Energy metrics
Energy Consumption Baysan et al. (2019); Bettoni et al. (2015); Cosgrove et al. (2018); Estrada-Gonzalez et al. (2020); Garza-Reyes et al. (2018); Jia et al.
(2017); Keskin et al. (2012); Muller et al. (2014); Müller et al. (2014); Rukmayadi et al. (2016); Schönemann et al. (2016); Sihag and
Sangwan (2019); Svensson and Paramonova (2017); Thanki and Thakkar (2016); Verma and Sharma (2019); Xie et al. (2018)
Energy Usage Baysan et al. (2019); Bettoni et al. (2015); Cosgrove et al. (2018); Estrada-Gonzalez et al. (2020); Garza-Reyes et al. (2018); Jia et al.
(2017); Keskin et al. (2012); Müller et al. (2014); Rukmayadi et al. (2016); Sihag and Sangwan (2019); Svensson and Paramonova (2017);
Thanki and Thakkar (2016); Verma and Sharma (2019); Xie et al. (2018)
Energy Saving Baysan et al. (2019); Bettoni et al. (2015); Cosgrove et al. (2018); Estrada-Gonzalez et al. (2020); Jia et al. (2017); Keskin et al. (2012);
Melsas and Rosin (2017); Muller et al. (2014); Rukmayadi et al. (2016); Sihag and Sangwan (2019); Svensson and Paramonova (2017);
Verma and Sharma (2019); Xie et al. (2018)
Energy Efficiency Baysan et al. (2019); Bettoni et al. (2015); Cosgrove et al. (2018); Keskin et al. (2012); Müller et al. (2014); Rukmayadi et al. (2016); Sihag
and Sangwan (2019); Svensson and Paramonova (2017); Thanki and Thakkar (2016); Verma and Sharma (2019); Xie et al. (2018)
Energy Cost Baysan et al. (2019); Bettoni et al. (2015); Cosgrove et al. (2018); Jia et al. (2017); Müller et al. (2014); Schönemann et al. (2016); Sihag
and Sangwan (2019); Thanki and Thakkar (2016); Xie et al. (2018)
Energy Management Baysan et al. (2019); Bettoni et al. (2015); Cosgrove et al. (2018); Garza-Reyes et al. (2018); Jia et al. (2017); Svensson and Paramonova
(2017)
Energy Input Cosgrove et al. (2018); Estrada-Gonzalez et al. (2020); Muller et al. (2014); Müller et al. (2014); Svensson and Paramonova (2017); Verma
and Sharma (2019)
Energy Conservation Baysan et al. (2019); Melsas and Rosin (2017); Svensson and Paramonova (2017); Verma and Sharma (2019)
Energy Value Bettoni et al. (2015); Sihag and Sangwan (2019); Verma and Sharma (2019)
Energy Utilisation Baysan et al. (2019); Keskin et al. (2012)
(continued on next page)

17
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

(continued )
References

Non-Value-Added Energy Bettoni et al. (2015); Verma and Sharma (2019)

Value-Added Energy Bettoni et al. (2015); Verma and Sharma (2019)
Energy Cost per Item Schönemann et al. (2016)
Energy Consumption for Holding Bettoni et al. (2015)
Function
Annual Constant Energy Consumption Melsas and Rosin (2017)
Reduction
Annual Variable Energy Consumption Melsas and Rosin (2017)
Reduction
Environmental and Production process metrics
Cycle Time Baysan et al. (2019); Cosgrove et al. (2018); Estrada-Gonzalez et al. (2020); Jia et al. (2017); Keskin et al. (2012); Melsas and Rosin (2017);
Muller et al. (2014); Müller et al. (2014); Rukmayadi et al. (2016); Schönemann et al. (2016); Sihag and Sangwan (2019); Svensson and
Paramonova (2017); Thanki and Thakkar (2016); Verma and Sharma (2019); Xie et al. (2018)
Utilisation/Uptime Baysan et al. (2019); Bettoni et al. (2015); Estrada-Gonzalez et al. (2020); Jia et al. (2017); Keskin et al. (2012); Muller et al. (2014); Müller
et al. (2014); Verma and Sharma (2019); Xie et al. (2018)
Changeover Time Baysan et al. (2019); Jia et al. (2017); Keskin et al. (2012); Müller et al. (2014); Rukmayadi et al. (2016); Sihag and Sangwan (2019);
Svensson and Paramonova (2017); Verma and Sharma (2019)
Lead Time Muller et al. (2014); Müller et al. (2014); Schönemann et al. (2016); 28; Thanki and Thakkar (2016); Verma and Sharma (2019)
Takt Time Baysan et al. (2019); Bettoni et al. (2015); Thanki and Thakkar (2016)
Value-Added Time/Availability Jia et al. (2017); Müller et al. (2014); Thanki and Thakkar (2016)
Greenhouse Gas Emissions Rukmayadi et al. (2016); Sihag and Sangwan (2019)
Waste generation rate Baysan et al. (2019); Rukmayadi et al. (2016)
Every Part Every (EPE) Jia et al. (2017)
Chemical materials Rukmayadi et al. (2016)
Noise level Rukmayadi et al. (2016)
Non-Value-Added Time Müller et al. (2014)
Power Required Bettoni et al. (2015)
Waiting Time Xie et al. (2018)
Water consumption Rukmayadi et al. (2016)
I4.0 technologies
Internet of Things (IoT) Balaji et al. (2020); Boonsothonsatit et al. (2020); Buer et al. (2018); Camgoz-Akdag et al. (2018); Fortuny-Santos et al. (2020); Huang
et al., 2019; Ilangakoon et al. (2021); Mayr et al. (2018); Pagliosa et al. (2021); Pasi et al. (2020); Phuong and Guidat (2018); Ramadan
et al. (2020); Tran et al. (2021); Wagner et al. (2018)
Big-Data Analytics (BDA) Buer et al. (2018); Fortuny-Santos et al. (2020); Ilangakoon et al. (2021); Mayr et al. (2018); Pagliosa et al. (2021); Pasi et al. (2020);
Phuong and Guidat (2018); Ramadan et al. (2020); Tran et al. (2021); Valamede and Akkari (2020); Wagner et al. (2018)
Industrial Robots Benzi et al. (2018); Boonsothonsatit et al. (2020); Buer et al. (2018); Fortuny-Santos et al. (2020); Ilangakoon et al. (2021); Mayr et al.
(2018); Pagliosa et al. (2021); Pasi et al. (2020); Phuong and Guidat (2018); Ramadan et al. (2020); Valamede and Akkari (2020)
Cyber-Physical Systems (CPS) Boonsothonsatit et al. (2020); Buer et al. (2018); Fortuny-Santos et al. (2020); Huang et al., 2019; Mayr et al. (2018); Pagliosa et al. (2021);
Ramadan et al. (2020); Tran et al. (2021); Wagner et al. (2018)
Cloud Computing Fortuny-Santos et al. (2020); Huang et al., 2019; Ilangakoon et al. (2021); Pagliosa et al. (2021); Pasi et al. (2020); Ramadan et al. (2020);
Valamede and Akkari (2020); Wagner et al. (2018)
Integration and Interoperability Fortuny-Santos et al. (2020); Huang et al., 2019; Pagliosa et al. (2021); Pasi et al. (2020); Phuong and Guidat (2018); Ramadan et al.
(2020); Wagner et al. (2018)
Additive Manufacturing (AM) Fortuny-Santos et al. (2020); Ilangakoon et al. (2021); Mayr et al. (2018); Pagliosa et al. (2021); Pasi et al. (2020); Valamede and Akkari
(2020)
Augmented Reality (AR) Ilangakoon et al. (2021); Mayr et al. (2018); Pagliosa et al. (2021); Valamede and Akkari (2020); Wagner et al. (2018)
Virtual Simulation (VS) Abideen and Mohamad (2021); Pagliosa et al. (2021); Valamede and Akkari (2020)
Virtual Reality (VR) Mayr et al. (2018); Tran et al. (2021); Wagner et al. (2018)
Digital Twin Mayr et al. (2018); Tran et al. (2021)
Cyber-Security Pasi et al. (2020); Valamede and Akkari (2020)
Fog Computing Ilangakoon et al. (2021)

Appendix B. Fuzzy Delphi responses

E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 P1 P2 P3 P4 P5 P6

M1 ER ER ER HR MR MR LR HR HR MR ER ER ER ER MR HR
M2 HR HR HR HR HR HR ER HR HR ER ER ER MR MR HR MR
M3 HR HR HR HR HR MR MR HR HR MR ER ER HR HR HR HR
M4 HR HR HR HR HR HR MR HR HR MR ER ER MR HR HR MR
M5 ER HR HR HR HR HR HR ER HR MR HR ER HR ER MR HR
M6 ER ER ER ER ER ER ER ER ER ER ER ER ER ER ER ER
M7 ER ER ER ER HR HR ER ER HR ER ER HR ER ER ER ER
M8 HR MR MR LR LR ER NR NR MR HR ER NR MR LR ER LR
M9 ER HR HR HR HR HR LR ER HR HR HR HR HR ER LR HR
M10 HR ER ER HR ER HR HR MR ER MR ER HR HR ER HR HR
M11 ER ER ER HR ER HR ER ER ER ER HR HR ER ER HR ER
M12 HR HR HR HR HR MR ER HR MR HR HR HR HR HR ER HR
M13 ER ER HR ER HR HR HR ER HR HR HR MR ER ER MR ER
M14 ER HR ER ER HR ER MR ER ER ER ER ER ER ER HR ER
M15 HR HR HR HR HR HR ER ER HR HR HR HR ER HR HR ER
M16 ER HR HR HR ER MR MR ER ER HR ER HR ER ER LR ER
M17 ER ER HR ER HR HR ER ER ER MR ER ER ER HR ER HR
(continued on next page)

18
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

(continued )
E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 P1 P2 P3 P4 P5 P6

M18 HR HR HR HR HR HR MR HR HR HR HR HR HR HR HR MR
M19 ER ER ER MR HR MR ER ER HR MR MR MR HR HR MR HR
M20 ER ER ER ER ER ER ER ER ER ER ER ER ER ER ER ER
M21 HR HR ER HR ER ER ER ER HR HR MR HR LR HR MR HR
M22 HR HR ER HR MR ER MR HR HR HR HR HR HR HR HR HR
M23 HR HR HR HR MR MR LR HR HR HR HR ER LR MR MR HR
M24 ER ER ER ER HR MR LR MR ER LR HR HR LR LR HR MR
E1: Energy Consumption; E2: Energy Usage; E3: Energy Cost; E4: Energy Input; E5: Energy Value; E6: Energy Utilisation; E7: Non-Value-Added Energy; E8: Value-
Added Energy; E9: Energy Cost per Item; E10: Energy Consumption per Defective Item; P1: Cycle Time; P2: Lead Time; P3: Greenhouse Gas Emissions; P4: Power
Required; P5: Waiting Time; P6: Water Consumption.

Appendix C. Initial direct relation matrix I – Respondent 1

P1 P3 P4 P5 P6

E1 EI EI EI HI EI
E2 EI EI EI HI EI
E3 HI EI EI HI EI
E4 EI EI EI HI EI
E5 EI EI EI EI EI
E6 EI EI EI EI EI
E9 EI EI EI EI EI
E10 EI EI EI EI EI
E1: Energy Consumption; E2: Energy Usage; E3: Energy Cost; E4: Energy Input; E5: Energy Value; E6: Energy
Utilisation; E9: Energy Cost per Item; E10: Energy Consumption per Defective Item; P1: Cycle Time; P3:
Greenhouse Gas Emissions; P4: Power Required; P5: Waiting Time; P6: Water Consumption.

Appendix D. Triangular Fuzzy numbers (a), Fuzzy defuzzification process (b, c) and direct relation matrix D (d) – Respondent 1

a P1 P3 P4 P5 P6

E1 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.5 0.7 0.9) (0.7 0.9 1)
E2 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.5 0.7 0.9) (0.7 0.9 1)
E3 (0.5 0.7 0.9) (0.7 0.9 1) (0.7 0.9 1) (0.5 0.7 0.9) (0.7 0.9 1)
E4 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.5 0.7 0.9) (0.7 0.9 1)
E5 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1)
E6 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1)
E9 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1)
E10 (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1) (0.7 0.9 1)
b fl fm fr fl fm fr fl fm fr fl fm fr fl fm fr
E1 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0 0 0) (0 0 0)
E2 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0 0 0) (0 0 0)
E3 (0 0 0) (0 0 0) (0 0 0) (0 0 0) (0 0 0)
E4 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0 0 0) (0 0 0)
E5 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0.4 0.4 0.2) (0 0 0)
E6 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0.4 0.4 0.2) (0 0 0)
E9 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0.4 0.4 0.2) (0 0 0)
E10 (0.4 0.4 0.2) (0 0 0) (0 0 0) (0.4 0.4 0.2) (0 0 0)
c lv rv lv rv lv rv lv rv lv rv
E1 0.40 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
E2 0.40 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
E3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
E4 0.40 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
E5 0.40 0.25 0.00 0.00 0.00 0.00 0.40 0.25 0.00 0.00
E6 0.40 0.25 0.00 0.00 0.00 0.00 0.40 0.25 0.00 0.00
E9 0.40 0.25 0.00 0.00 0.00 0.00 0.40 0.25 0.00 0.00
E10 0.40 0.25 0.00 0.00 0.00 0.00 0.40 0.25 0.00 0.00
d x x x x x
E1 0.356 0.000 0.000 0.000 0.000
E2 0.356 0.000 0.000 0.000 0.000
E3 0.000 0.000 0.000 0.000 0.000
E4 0.356 0.000 0.000 0.000 0.000
E5 0.356 0.000 0.000 0.356 0.000
E6 0.356 0.000 0.000 0.356 0.000
E9 0.356 0.000 0.000 0.356 0.000
E10 0.356 0.000 0.000 0.356 0.000
E1: Energy Consumption; E2: Energy Usage; E3: Energy Cost; E4: Energy Input; E5: Energy Value; E6: Energy Utilisation; E9: Energy Cost per Item; E10: Energy
Consumption per Defective Item; P1: Cycle Time; P3: Greenhouse Gas Emissions; P4: Power Required; P5: Waiting Time; P6: Water Consumption.

19
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Appendix E. Direct relation matrix G for energy metrics on environmental and production process metrics

P1 P3 P4 P5 P6

E1 0.254 0.272 0.235 0.189 0.207

E2 0.229 0.230 0.206 0.179 0.173
E3 0.175 0.200 0.191 0.163 0.162
E4 0.125 0.231 0.205 0.137 0.114
E5 0.175 0.229 0.189 0.210 0.185
E6 0.257 0.275 0.197 0.292 0.207
E9 0.223 0.225 0.211 0.246 0.183
E10 0.152 0.169 0.152 0.230 0.181
E1: Energy Consumption; E2: Energy Usage; E3: Energy Cost; E4: Energy Input; E5: Energy Value; E6: Energy Utilisation; E9:
Energy Cost per Item; E10: Energy Consumption per Defective Item; P1: Cycle Time; P3: Greenhouse Gas Emissions; P4: Power
Required; P5: Waiting Time; P6: Water Consumption.

References supply chain management—a framework headed toward the 2030 agenda.
Sustainability 14 (2), 830. https://doi.org/10.3390/su14020830.
Camgoz-Akdag, H., Beldek, T., Konyalioglu, A.K., 2018. Process improvement in A
Abdulmalek, F.A., Rajgopal, J., 2007. Analyzing the benefits of lean manufacturing and
radiology department with value stream mapping and its linkage to industry 4.0.
value stream mapping via simulation: a process sector case study. Int. J. Prod. Econ.
IIOAB J. 9 (6), 36–41. WE-Emerging Sources Citation Index (ESCI).
107 (1), 223–236. https://doi.org/10.1016/j.ijpe.2006.09.009.
Chang, A.-Y., Cheng, Y.-T., 2019. Analysis model of the sustainability development of
Abideen, A., Mohamad, F.B., 2021. Improving the performance of a Malaysian
manufacturing small and medium- sized enterprises in Taiwan. J. Clean. Prod. 207,
pharmaceutical warehouse supply chain by integrating value stream mapping and
458–473. https://doi.org/10.1016/j.jclepro.2018.10.025.
discrete event simulation. J. Model. Manag. 16 (1), 70–102. https://doi.org/
Chang, B., Chang, C.-W., Wu, C.-H., 2011. Fuzzy DEMATEL method for developing
10.1108/JM2-07-2019-0159 WE - Emerging Sources Citation Index (ESCI).
supplier selection criteria. Expert Syst. Appl. 38 (3), 1850–1858. https://doi.org/
Adenuga, O.T., Mpofu, K., Boitumelo, R.I., 2019. Energy efficiency analysis modelling
10.1016/j.eswa.2010.07.114.
system for manufacturing in the context of industry 4.0. Procedia CIRP 80, 735–740.
Chang, P.-T., Huang, L.-C., Lin, H.-J., 2000. The fuzzy Delphi method via fuzzy statistics
https://doi.org/10.1016/j.procir.2019.01.002.
and membership function fitting and an application to the human resources. Fuzzy
Akbaba, M., 1999. Energy conservation by using energy efficient electric motors. Appl.
Set Syst. 112 (3), 511–520. https://doi.org/10.1016/S0165-0114(98)00067-0.
Energy 64 (1–4), 149–158. https://doi.org/10.1016/S0306-2619(99)00051-3.
Chen, G.Q., Wu, X.D., Guo, J., Meng, J., Li, C., 2019. Global overview for energy use of
Andreadis, E., Garza-Reyes, J.A., Kumar, V., 2017. Towards a conceptual framework for
the world economy: household-consumption-based accounting based on the world
value stream mapping (VSM) implementation: an investigation of managerial
input-output database (WIOD). Energy Econ. 81, 835–847. https://doi.org/10.1016/
factors. Int. J. Prod. Res. 55 (23), 7073–7095. https://doi.org/10.1080/
j.eneco.2019.05.019.
00207543.2017.1347302.
Cheung, W.W.L., Pitcher, T.J., Pauly, D., 2005. A fuzzy logic expert system to estimate
Arksey, H., O’Malley, L., 2005. Scoping studies: towards a methodological framework.
intrinsic extinction vulnerabilities of marine fishes to fishing. Biol. Conserv. 124 (1),
Int. J. Soc. Res. Methodol. 8 (1), 19–32. https://doi.org/10.1080/
97–111. https://doi.org/10.1016/j.biocon.2005.01.017.
1364557032000119616.
Cosgrove, J., Rivas Duarte, M.-J., Littlewood, J., Wilgeroth, P., 2018. An energy mapping
Ashtiani Araghi, Z., Vosoughifar, H., 2023. Modified BIM processes considering
methodology to reduce energy consumption in manufacturing operations. Proc. IME
safety–quality index for precast concrete construction. ASCE-ASME J. Risk.
B J. Eng. Manufact. 232 (10), 1731–1740. https://doi.org/10.1177/
Uncertain. Eng. Sys. Part A.: Civ. Eng. 9 (1) https://doi.org/10.1061/
0954405416673101.
AJRUA6.0001274.
Crowe, S., Cresswell, K., Robertson, A., Huby, G., Avery, A., Sheikh, A., 2011. The case
Ates, S.A., Durakbasa, N.M., 2012. Evaluation of corporate energy management practices
study approach. BMC Med. Res. Methodol. 11 (1), 100. https://doi.org/10.1186/
of energy intensive industries in Turkey. Energy 45 (1), 81–91. https://doi.org/
1471-2288-11-100.
10.1016/j.energy.2012.03.032.
Danese, P., Manfè, V., Romano, P., 2018. A systematic literature review on recent lean
Balaji, V., Venkumar, P., Sabitha, M.S., Amuthaguka, D., 2020. DVSMS: dynamic value
research: state-of-the-art and future directions. Int. J. Manag. Rev. 20 (2), 579–605.
stream mapping solution by applying IIoT. Sadhana - Acad. Proc. Eng. Sci. 45 (1)
https://doi.org/10.1111/ijmr.12156.
https://doi.org/10.1007/s12046-019-1251-5.
de Mattos Nascimento, D.L., Mury Nepomuceno, R., Caiado, R.G.G., Maqueira, J.M.,
Baysan, S., Kabadurmus, O., Cevikcan, E., Satoglu, S.I., Durmusoglu, M.B., 2019.
Moyano-Fuentes, J., Garza-Reyes, J.A., 2022. A sustainable circular 3D printing
A simulation-based methodology for the analysis of the effect of lean tools on energy
model for recycling metal scrap in the automotive industry. J. Manuf. Technol.
efficiency: an application in power distribution industry. J. Clean. Prod. 211,
Manag. 33 (5), 876–892. https://doi.org/10.1108/JMTM-10-2021-0391.
895–908. https://doi.org/10.1016/j.jclepro.2018.11.217.
del Caño, A., de la Cruz, M.P., 2002. Integrated methodology for project risk
Bazurto, A.J., Quispe, E.C., Mendoza, R.C., 2017. Causes and failures classification of
management. J. Construct. Eng. Manag. 128 (6), 473–485. https://doi.org/10.1061/
industrial electric motor. In: Proceedings of the 2016 IEEE ANDESCON. ANDESCON.
(ASCE)0733-9364(2002)128:6(473).
https://doi.org/10.1109/ANDESCON.2016.7836190, 2016.
Di Pasquale, V., Miranda, S., Neumann, W.P., 2020. Ageing and human-system errors in
Benzi, F., Clava, R., Bassi, E., 2018. Increasing the Value Stream Mapping Potential in an
manufacturing: a scoping review. Int. J. Prod. Res. 58 (15), 4716–4740. https://doi.
Industrial Process, with a Dynamic Model, Based on Data from an Industrial Ethernet
org/10.1080/00207543.2020.1773561.
Bus. 2018 ECEEE Industrial Summer Study On Industrial Efficiency: Leading The Low-
Dombrowski, U., Mielke, T. (Eds.), 2015. Ganzheitliche Produktionssysteme. Springer
Carbon Transition, 2018-June, pp. 189–195. https://www.scopus.com/inward/rec
Berlin Heidelberg. https://doi.org/10.1007/978-3-662-46164-8.
ord.uri?eid=2-s2.0-85049915826&partnerID=40&md5=a00cbb64d41a0adb8c
Estrada-Gonzalez, I.E., Taboada-Gonzalez, P.A., Guerrero-Garcia-Rojas, H., Marquez-
fd3eb0b576cced.
Benavides, L., 2020. Decreasing the environmental impact in an egg-producing farm
Bettoni, L., Mazzoldi, L., Ferretti, I., Zavanella, L., Zanoni, S., 2015. Integrated energy
through the application of LCA and lean tools. App. Sci. Basel. 10 (4) https://doi.
value analysis: a new approach. In: Umeda, S., Nakano, M., Mizuyama, H.,
org/10.3390/app10041352.
Hibino, H., Kiritsis, D., VonCieminski, G. (Eds.), Advances In Production Management
Falatoonitoosi, E., Leman, Z., Sorooshian, S., Salimi, M., 2013. Decision-making trial and
Systems: Innovative Production Management towards Sustainable Growth (Amps 2015),
evaluation laboratory. Res. J. Appl. Sci. Eng. Technol. 5 (13), 3476–3480. https://
PT I (Vol. 459, Issue IFIP WG 5.7 International Conference on Advances in
doi.org/10.19026/rjaset.5.4475.
Production Management Systems (APMS), pp. 670–679. https://doi.org/10.1007/
Fernández-Cerero, D., Fernández-Montes, A., Javier Ortega, F., Jakóbik, A., Widlak, A.,
978-3-319-22756-6_82.
Fernandez-Cerero, D., Fernandez-Montes, A., Ortega, F.J., Jakobik, A., Widlak, A.,
Boonsothonsatit, G., Tonchiangsai, K., Choowitsakunlert, S., 2020. Value stream
2020. Sphere: simulator of edge infrastructures for the optimization of performance
mapping-based logistics 4.0 readiness for Thailand automotive-Part Manufacturers.
and resources energy consumption. Simulat. Model. Pract. Theor. 101 https://doi.
In: 2nd International Conference on Management Science and Industrial
org/10.1016/j.simpat.2019.101966.
Engineering. MSIE, pp. 309–313. https://doi.org/10.1145/3396743.3396745, 2020.
Ferreira, W. de P., Armellini, F., Santa-Eulalia, L. A. de, Thomasset-Laperrière, V., 2022.
Borowski, P.F., 2022. Mitigating climate change and the development of green energy
Extending the lean value stream mapping to the context of Industry 4.0: an agent-
versus a return to fossil fuels due to the energy crisis in 2022. Energies 15 (24), 9289.
based technology approach. J. Manuf. Syst. 63, 1–14. https://doi.org/10.1016/j.
https://doi.org/10.3390/en15249289.
jmsy.2022.02.002.
Buer, S.-V., Strandhagen, J.O., Chan, F.T.S., 2018. The link between industry 4.0 and
Fortuny-Santos, J., López, P.R.-D.-A., Luján-Blanco, I., Chen, P.-K., 2020. Assessing the
lean manufacturing: mapping current research and establishing a research agenda.
synergies between lean manufacturing and Industry 4.0. Direccion y Organizacion
Int. J. Prod. Res. 56 (8), 2924–2940. https://doi.org/10.1080/
71, 71–86. https://doi.org/10.37610/dyo.v0i71.579.
00207543.2018.1442945.
Gabriele, P.D., Treinta, F.T., De Farias Filho, J.R., De Souza, M.C., Tschaffon, P.B.,
Caiado, R.G.G., Scavarda, L.F., Azevedo, B.D., de Mattos Nascimento, D.L., Quelhas, O.L.
Brantes, S.R., 2012. Sustentabilidade e vantagem competitiva estratégica: um estudo
G., 2022. Challenges and benefits of sustainable industry 4.0 for operations and
exploratório e bibliométrico. Rev. Produção Online 12 (3), 729. https://doi.org/
10.14488/1676-1901.v12i3.975.

20
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Garcia-Buendia, N., Kristensen, T.B., Moyano-Fuentes, J., Maqueira-Marín, J.M., 2022. Marin-Garcia, J.A., Vidal-Carreras, P.I., Garcia-Sabater, J.J., 2021. The role of value
Performance measurement of lean supply chain management: a balanced scorecard stream mapping in healthcare services: a scoping review. Int. J. Environ. Res. Publ.
proposal. Prod. Plann. Control 1–21. https://doi.org/10.1080/ Health 18 (3), 951. https://doi.org/10.3390/ijerph18030951.
09537287.2022.2114960. Mayr, A., Weigelt, M., Kühl, A., Grimm, S., Erll, A., Potzel, M., Franke, J., 2018. Lean 4.0-
Garza-Reyes, J.A., Kumar, V., Chaikittisilp, S., Tan, K.H., 2018. The effect of lean A conceptual conjunction of lean management and Industry 4.0. In: K, T., W, L.,
methods and tools on the environmental performance of manufacturing J, W., W, X.V. (Eds.), 51st CIRP Conference on Manufacturing Systems, CIRP CMS
organisations. Int. J. Prod. Econ. 200, 170–180. https://doi.org/10.1016/j. 2018, vol. 72. Elsevier B.V, pp. 622–628. https://doi.org/10.1016/j.
ijpe.2018.03.030. procir.2018.03.292.
Garza-Reyes, Arturo, Jose, 2015. Lean and green – a systematic review of the state of the Melsas, R., Rosin, A., 2017. Use of value stream mapping for evaluation of load
art literature. J. Clean. Prod. 102, 18–29. https://doi.org/10.1016/j. conservation and peak clipping possibilities. In: 17th IEEE International Conference
jclepro.2015.04.064. on Environment and Electrical Engineering and 2017 1st IEEE Industrial and
Gholami, H., Jamil, N., Zakuan, N., Mat Saman, M.Z., Sharif, S., Awang, S.R., Commercial Power Systems Europe, EEEIC/I and CPS Europe 2017. https://doi.org/
Sulaiman, Z., 2019. Social value stream mapping (Socio-VSM): methodology to 10.1109/EEEIC.2017.7977549.
societal sustainability visualization and assessment in the manufacturing system. Meudt, T., Metternich, J., Abele, E., 2017. Value stream mapping 4.0: holistic
IEEE Access 7, 131638–131648. https://doi.org/10.1109/ACCESS.2019.2940957. examination of value stream and information logistics in production. CIRP Annals 66
Heravi, G., Rostami, M., Kebria, M.F., 2020. Energy consumption and carbon emissions (1), 413–416. https://doi.org/10.1016/j.cirp.2017.04.005.
assessment of integrated production and erection of buildings’ pre-fabricated steel Moktadir, M.A., Rahman, T., Ali, S.M., Nahar, N., Paul, S.K., 2020. Examining barriers to
frames using lean techniques. J. Clean. Prod. 253, 120045 https://doi.org/10.1016/ reverse logistics practices in the leather footwear industry. Ann. Oper. Res. 293 (2),
j.jclepro.2020.120045. 715–746. https://doi.org/10.1007/s10479-019-03449-y.
Hermann, M., Pentek, T., Otto, B., 2015. Design Principles for Industrie 4.0 Scenarios: A Morgan, D., 1997. Focus Groups as Qualitative Research. SAGE Publications, Inc.
Literature Review. https://doi.org/10.13140/RG.2.2.29269.22248. https://doi.org/10.4135/9781412984287.
Hsu, Hsi-Mei, Chen, Chen-Tung, 1996. Aggregation of fuzzy opinions under group Morgan, D.L., 1996. Focus groups. Annu. Rev. Sociol. 22 (1), 129–152. https://doi.org/
decision making. Fuzzy Set Syst. 79 (3), 279–285. https://doi.org/10.1016/0165- 10.1146/annurev.soc.22.1.129.
0114(95)00185-9. Muller, E., Schillig, R., Stock, T., Schmeiler, M., Müller, E., Schillig, R., Stock, T.,
Huang, Z., Kim, J., Sadri, A., Dowey, S., Dargusch, M.S., 2019a. Industry 4.0: Schmeiler, M., 2014. Improvement of injection moulding processes by using dual
development of a multi-agent system for dynamic value stream mapping in SMEs. energy signatures. In: ElMaraghy, H. (Ed.), 47th CIRP Conference on Manufacturing
J. Manuf. Syst. 52, 1–12. https://doi.org/10.1016/j.jmsy.2019.05.001. Systems, CMS 2014. Elsevier B.V, pp. 704–709. https://doi.org/10.1016/j.
Huang, Z., Kim, J., Sadri, A., Dowey, S., Dargusch, M.S., 2019b. Industry 4.0: procir.2014.01.110. Vol. 17, Issue 47th CIRP Conference on Manufacturing Systems.
development of a multi-agent system for dynamic value stream mapping in SMEs. Müller, E., Stock, T., Schillig, R., 2014. A method to generate energy value-streams in
J. Manuf. Syst. 52, 1–12. https://doi.org/10.1016/j.jmsy.2019.05.001. production and logistics in respect of time- and energy-consumption. J. Inst. Eng.
IEA, 2020. Tracking Industry 2020. International Energy Agency. https://www.iea.org/r Prod. 8 (1–2), 243–251. https://doi.org/10.1007/s11740-013-0516-9.
eports/tracking-industry-2020. Nascimento, D.L. de M., Quelhas, O.L.G., Moyano-Fuentes, J., Tortorella, G.L.,
IEA, 2021a. Global Energy-Related CO2 Emissions by Sector. International Energy Maqueira, J.M., 2022. Circular value stream mapping 4.0: proposed general model
Agency. https://www.iea.org/data-and-statistics/charts/global-energy-related- and application to a digital 3D printing recycling factory. Sustain. Prod. Consum. 34,
co2-emissions-by-sector. 600–612. https://doi.org/10.1016/j.spc.2022.10.012.
IEA, 2021b. Key World Energy Statistics 2021. International Energy Agency. htt Nounou, A., Jaber, H., Aydin, R., 2022. A cyber-physical system architecture based on
ps://www.iea.org/reports/key-world-energy-statistics-2021/final-consumption. lean principles for managing industry 4.0 setups. Int. J. Comput. Integrated Manuf.
Ilangakoon, T.S., Weerabahu, S.K., Samaranayake, P., Wickramarachchi, R., 2021. 35 (8), 890–908. https://doi.org/10.1080/0951192X.2022.2027016.
Adoption of Industry 4.0 and lean concepts in hospitals for healthcare operational Núñez-Merino, M., Maqueira-Marín, J.M., Moyano-Fuentes, J., Martínez-Jurado, P.J.,
performance improvement. Int. J. Prod. Perform. Manag. 71 (6), 2188–2213. 2020. Information and digital technologies of Industry 4.0 and Lean supply chain
https://doi.org/10.1108/IJPPM-12-2020-0654. management: a systematic literature review. Int. J. Prod. Res. 58 (16), 5034–5061.
Ishikawa, A., 1993. The new fuzzy Delphi methods: economization of GDS (group https://doi.org/10.1080/00207543.2020.1743896.
decision support). 26th Hawaii Int. Conf. Syst. Sci, HICSS 4, 255–264. https://doi. Okoli, C., Pawlowski, S.D., 2004. The Delphi method as a research tool: an example,
org/10.1109/HICSS.1993.284190, 1993. design considerations and applications. Inf. Manag. 42 (1), 15–29. https://doi.org/
Ishikawa, Akira, Amagasa, M., Shiga, T., Tomizawa, G., Tatsuta, R., Mieno, H., 1993. The 10.1016/j.im.2003.11.002.
max-min Delphi method and fuzzy Delphi method via fuzzy integration. Fuzzy Set Oliveira, M., Freitas, H.M.R., 1998. Focus Group - pesquisa qualitativa: resgatando a
Syst. 55 (3), 241–253. https://doi.org/10.1016/0165-0114(93)90251-C. teoria, instrumentalizando o seu planejamento. Rev. Adm. 33 (3), 83–91.
Jamil, N., Gholami, H., Saman, M.Z.M., Streimikiene, D., Sharif, S., Zakuan, N., 2020. Pagliosa, M., Tortorella, G., Ferreira, J.C.E., 2021. Industry 4.0 and Lean Manufacturing:
DMAIC-based approach to sustainable value stream mapping: towards a sustainable a systematic literature review and future research directions. J. Manuf. Technol.
manufacturing system. Econ. Res.-Ekonomska Istraživanja 33 (1), 331–360. https:// Manag. 32 (3), 543–569. https://doi.org/10.1108/JMTM-12-2018-0446.
doi.org/10.1080/1331677X.2020.1715236. Palm, J., Thollander, P., 2010. An interdisciplinary perspective on industrial energy
Jia, S., Yuan, Q., Lv, J., Liu, Y., Ren, D., Zhang, Z., 2017. Therblig-embedded value efficiency. Appl. Energy 87 (10), 3255–3261. https://doi.org/10.1016/j.
stream mapping method for lean energy machining. Energy 138, 1081–1098. apenergy.2010.04.019.
https://doi.org/10.1016/j.energy.2017.07.120. Pasi, B.N., Mahajan, S.K., Rane, S.B., 2020. Redesigning of smart manufacturing system
Jirkovsky, V., Obitko, M., Marik, V., 2017. Understanding data heterogeneity in the based on iot: perspective of disruptive innovations of industry 4.0 paradigm. Int. J.
context of cyber-physical systems integration. IEEE Trans. Ind. Inf. 13 (2), 660–667. Mech. Prod. Eng. Res. Dev. 10 (3), 727–746. https://doi.org/10.24247/
https://doi.org/10.1109/TII.2016.2596101. ijmperdjun202067.
Johansson, M.T., Thollander, P., 2018. A review of barriers to and driving forces for Pereira, A., Simonetto, E. de O., 2018. Indústria 4.0: conceitos e perspectivas para o
improved energy efficiency in Swedish industry – recommendations for successful Brasil. Revista Da Universidade Vale Do Rio Verde 16 (1).
in-house energy management. Renew. Sustain. Energy Rev. 82, 618–628. https:// Phuong, N.A., Guidat, T., 2018. Sustainable value stream mapping and technologies of
doi.org/10.1016/j.rser.2017.09.052. Industry 4.0 in manufacturing process reconfiguration: a case study in an apparel
Keskin, C., Asan, U., Kayakutlu, G., 2012. Value stream maps for industrial energy company. In: 2018 IEEE International Conference on Service Operations and
efficiency. In: 2012 Portland International Conference on Management of Logistics, and Informatics, SOLI 2018, pp. 85–90. https://doi.org/10.1109/
Engineering and Technology - Technology Management for Emerging Technologies, SOLI.2018.8476750.
pp. 2824–2831. https://doi.org/10.1007/978-1-4471-5143-2_17. PICMET’12, 129. Qarnain, S.S., Muthuvel, S., Bathrinath, S., Saravanasankar, S., 2021. Analyzing factors
Kontio, J., Bragge, J., Lehtola, L., 2008. In: Shull, F., Singer, J., Sjøberg, D.I.K. (Eds.), The in emerging computer technologies favoring energy conservation of building sector.
Focus Group Method as an Empirical Tool in Software Engineering BT - Guide to Mater. Today: Proc. 45, 1290–1293. https://doi.org/10.1016/j.matpr.2020.05.121.
Advanced Empirical Software Engineering, 93–116). Springer London. https://doi. Quelhas, A.D., Filho, J.R.F., Neto, J.V., Pereira, V., 2019. Model to measure adherence of
org/10.1007/978-1-84800-044-5_4. culture, climate, and organizational behavior in a construction company. J. Manag.
Krueger, R.A., 1994. Focus Groups: A Practical Guide for Applied Research, second ed. Eng. 35 (4) https://doi.org/10.1061/(ASCE)ME.1943-5479.0000688.
SAGE Publications Ltd. Ramadan, M., Salah, B., Othman, M., Ayubali, A.A., 2020. Industry 4.0-based real-time
Lee, C.-H., Wu, K.-J., Tseng, M.-L., 2018. Resource management practice through eco- scheduling and dispatching in lean manufacturing systems. Sustainability 12 (6).
innovation toward sustainable development using qualitative information and https://doi.org/10.3390/su12062272.
quantitative data. J. Clean. Prod. 202, 120–129. https://doi.org/10.1016/j. Rojas, E.M., Dossick, C.S., 2008. Developing a state-of-the-art facility to support
jclepro.2018.08.058. construction research and education: a case study. J. Prof. Issues Eng. Educ. Pract.
Lee, J.K.Y., Gholami, H., Saman, M.Z.M., Ngadiman, N.H.A., Zakuan, N., Mahmood, S., 134 (1), 67–74. https://doi.org/10.1061/(ASCE)1052-3928(2008)134:1(67).
Omain, S.Z., 2021. Sustainability-oriented application of value stream mapping: a Rukmayadi, D., Marimin, Haris, U., Yani, M., 2016. Rubber agro-industry green logistic
review and classification. IEEE Access 9, 68414–68434. https://doi.org/10.1109/ conceptual model. Int. J. Supply Chain Manag. 5 (3), 192–204. https://www.scopus.
ACCESS.2021.3077570. com/inward/record.uri?eid=2-s2.0-84990059353&partnerID=40&md5=ebd0767c
Lucko, G., Rojas, E.M., 2010. Research validation: challenges and opportunities in the e893c6d07db36dc8841f779f.
construction domain. J. Construct. Eng. Manag. 136 (1), 127–135. https://doi.org/ Saieg, P., Sotelino, E.D., Nascimento, D., Caiado, R.G.G., 2018. Interactions of building
10.1061/(ASCE)CO.1943-7862.0000025. information modeling, lean and sustainability on the architectural, engineering and
Machado, E., Scavarda, L.F., Caiado, R.G.G., Thomé, A.M.T., 2021. Barriers and enablers construction industry: a systematic review. J. Clean. Prod. 174, 788–806. https://
for the integration of industry 4.0 and sustainability in supply chains of MSMEs. doi.org/10.1016/j.jclepro.2017.11.030.
Sustainability 13 (21), 11664. https://doi.org/10.3390/su132111664.

21
L. Fontoura et al. Journal of Cleaner Production 419 (2023) 137871

Saktiawan, B., Juan Suam Toro, M., Saputro, N., 2022. The impact of the Russia- Tsai, F.M., Bui, T.-D., Tseng, M.-L., Wu, K.-J., Chiu, A.S.F., 2020. A performance
Ukrainian war on green energy financing in Europe. IOP Conf. Ser. Earth Environ. assessment approach for integrated solid waste management using a sustainable
Sci. 1114 (1), 012066 https://doi.org/10.1088/1755-1315/1114/1/012066. balanced scorecard approach. J. Clean. Prod. 251, 119740 https://doi.org/10.1016/
Saligkaras, D., Papageorgiou, V.E., 2022. On the detection of patterns in electricity prices j.jclepro.2019.119740.
across European countries: an unsupervised machine learning approach. AIMS Valamede, L.S., Akkari, A.C.S., 2020. Lean 4.0: a new holistic approach for the
Energy 10 (6), 1146–1164. https://doi.org/10.3934/energy.2022054. integration of lean manufacturing tools and digital technologies. Int. J. Math. Eng.
Saunders, M.N.K., Lewis, P., Thornhill, A., 2012. Research Methods for Business Students Manag. Sci. 5 (5), 854–868. https://doi.org/10.33889/IJMEMS.2020.5.5.066.
(Sixth). Pearson Education Limited. Verma, N., Sharma, V., 2019. A literature review on energy value stream mapping
Schönemann, M., Kurle, D., Herrmann, C., Thiede, S., 2016. Multi-product EVSM (EVSM). Int. J. Adv. Sci. Technol. 27 (1), 1–8. https://www.scopus.com/inward/rec
simulation. In: T, R. (Ed.), 48th CIRP International Conference on Manufacturing ord.uri?eid=2-s2.0-85081190350&partnerID=40&md5=4ea2e23788ad9dedcc852
Systems, CIRP CMS 2015, vol. 41. Elsevier B.V, pp. 334–339. https://doi.org/ 0f875ed2a0f.
10.1016/j.procir.2015.10.012. Wagner, T., Herrmann, C., Thiede, S., 2018. Identifying target oriented Industrie 4.0
Schumacher, S., Bildstein, A., Bauernhansl, T., 2020. The impact of the digital potentials in lean automotive electronics value streams. In: K, T., W, L., J, W., W, X.
transformation on lean production systems. Procedia CIRP 93, 783–788. https://doi. V. (Eds.), 51st CIRP Conference on Manufacturing Systems, vol. 72. Elsevier B.V,
org/10.1016/j.procir.2020.03.066. pp. 1003–1008. https://doi.org/10.1016/j.procir.2018.03.003. CIRP CMS 2018.
Siddaway, A., 2014. What is a systematic literature review and how do I do one. Univ. Wang, F.-K., Rahardjo, B., Rovira, P.R., 2022. Lean six sigma with value stream mapping
Stirling. 1 (1), 1–13. in industry 4.0 for human-centered workstation design. Sustainability 14 (17),
Sihag, N., Sangwan, K.S., 2019. An improved micro analysis-based energy consumption 11020. https://doi.org/10.3390/su141711020.
and carbon emissions modeling approach for a milling center. Int. J. Adv. Des. Webster, J., Watson, R.T., 2002. Analyzing the past to prepare for the future: writing a
Manuf. Technol. 104 (1–4), 705–721. https://doi.org/10.1007/s00170-019-03807- literature review. MIS Q. 26 (2) xiii–xxiii. http://www.jstor.org/stable/4132319.
x. Wilhelm, J., Petzoldt, C., Beinke, T., Freitag, M., 2021. Review of digital twin-based
Stake, R.E., 1995. The Art of Case Study Research. Sage Publications Ltd. interaction in smart manufacturing: enabling cyber-physical systems for human-
Sulaiman, M.A., Ito, T., Rahman, M.A.A., Ishak, A., Salleh, M. R. Bin, Mohamad, E., 2019. machine interaction. Int. J. Comput. Integrated Manuf. 34 (10), 1031–1048. https://
Cleaner production value stream mapping at a chromium plating plant: a case study. doi.org/10.1080/0951192X.2021.1963482.
Int. J. Agile Syst. Manag. 12 (3), 245. https://doi.org/10.1504/ Wu, K.-J., Liao, C.-J., Chen, C.-C., Lin, Y., Tsai, C.F.M., 2016. Exploring eco-innovation in
IJASM.2019.10022800. dynamic organizational capability under incomplete information in the Taiwanese
Svensson, A., Paramonova, S., 2017. An analytical model for identifying and addressing lighting industry. Int. J. Prod. Econ. 181, 419–440. https://doi.org/10.1016/j.
energy efficiency improvement opportunities in industrial production systems – ijpe.2015.10.007.
model development and testing experiences from Sweden. J. Clean. Prod. 142, Wu, X.D., Guo, J.L., Ji, X., Chen, G.Q., 2019. Energy use in world economy from
2407–2422. https://doi.org/10.1016/j.jclepro.2016.11.034. household-consumption-based perspective. Energy Pol. 127, 287–298. https://doi.
Thanki, S.J., Thakkar, J.J., 2016. Value–value load diagram: a graphical tool for org/10.1016/j.enpol.2018.12.005.
lean–green performance assessment. Prod. Plann. Control 27 (15), 1280–1297. Xie, H., Chowdhury, M.M., Issa, R.R.A., Shi, W., 2018. Simulation of dynamic energy
https://doi.org/10.1080/09537287.2016.1220647. consumption in modular construction manufacturing processes. J. Architect. Eng. 24
Tong, A., Sainsbury, P., Craig, J., 2007. Consolidated criteria for reporting qualitative (1) https://doi.org/10.1061/(ASCE)AE.1943-5568.0000289.
research (COREQ): a 32-item checklist for interviews and focus groups. Int. J. Qual. Xu, L.D., Xu, E.L., Li, L., 2018. Industry 4.0: state of the art and future trends. Int. J. Prod.
Health Care 19 (6), 349–357. https://doi.org/10.1093/intqhc/mzm042. Res. 56 (8), 2941–2962. https://doi.org/10.1080/00207543.2018.1444806.
Tortorella, G.L., Fogliatto, F.S., Tlapa Mendoza, D., Pepper, M., Capurro, D., 2022. Yin, R.K., 2009. Case Study Research: Design and Method, fourth ed. SAGE Publications
Digital transformation of health services: a value stream-oriented approach. Int. J. Ltd.
Prod. Res. 1–15. https://doi.org/10.1080/00207543.2022.2048115. Yu, A.T.W., Shen, Q., Kelly, J., Hunter, K., 2006. Investigation of critical success factors
Tran, T.A., Ruppert, T., Abonyi, J., 2021. Indoor positioning systems can revolutionise in construction project briefing by way of content analysis. J. Construct. Eng. Manag.
digital lean. App. Sci. Basel. 11 (11) https://doi.org/10.3390/app11115291. 132 (11), 1178–1186. https://doi.org/10.1061/(ASCE)0733-9364(2006)132:11
Tripathi, V., Chattopadhyaya, S., Bhadauria, A., Sharma, S., Li, C., Pimenov, D.Y., (1178).
Giasin, K., Singh, S., Gautam, G.D., 2021. An agile system to enhance productivity Zhu, X.-Y., Zhang, H., Jiang, Z.-G., 2020. Application of green-modified value stream
through a modified value stream mapping approach in industry 4.0: a novel mapping to integrate and implement lean and green practices: a case study. Int. J.
approach. Sustainability 13 (21), 11997. https://doi.org/10.3390/su132111997. Comput. Integrated Manuf. 33 (7), 716–731. https://doi.org/10.1080/
0951192X.2019.1667028.

22