Heliyon 10 (2024) e26037

Contents lists available at ScienceDirect

Heliyon
journal homepage: www.cell.com/heliyon

Research article

Incorporating inflation rate in construction projects cost:

Forecasting model
Muhammad Ali Musarat a, b, Wesam Salah Alaloul a, *, M.S. Liew a
a
Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610, Perak, Malaysia
b
Offshore Engineering Centre, Institute of Autonomous System, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 32610, Perak, Malaysia

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

Keywords: Over time, the change in the inflation rate causes cost overruns by deviating the prices of goods
Cost overrun and services in construction projects that require practitioners to make budgeting revisions.
Inflation Hence, this study aims to develop a construction rates forecasting model that can incorporate the
Construction rates
changing impact of the inflation rate on construction rates and predict the prices in a particular
Forecasting model
year, which can be adjusted when developing the Bill of Quantities. Following the time series
Time series
Forecasting analysis standards, a mathematical model was developed using MATLAB for forecasting. Con­
struction rates, building prices, labour wages and machinery rates were forecasted from 2020 to
2025 based on the data collected from 2013 to 2019. Akaike information criterion was used to
validate the self-developed construction rate forecasting model. It was revealed that the model
yielded better results when the construction rates were compared with the autoregressive inte­
grated moving average time series model results. The rates forecasting model may be used for any
construction project where rates are affected by the inflation effect.

1. Introduction

The construction industry is one of the greatest contributors to society’s development; however, it continues to struggle under one
of the biggest constraints, i.e., cost overrun [1]. Cost overrun phenomena charge additional money and are a strong project failure
indicator [2]. Generally, a cost overrun on project completion occurs due to negligence in handling the associated risks at the
implementation level [3]. Cost overrun risk is always present in every construction project and continues to concern the stakeholders.
The rate of cost overrun varies in project type; however, an estimated range lies from 21% to 55% [4–6]. The major impact of cost
overrun can be attributed to the price changes of essential resources, i.e., materials, labour and machinery. Hence, accurate mea­
surement of these resources in project cost is important [7–9], and these resources must be handled in the initial planning phase [10].
Understanding the future behaviour of the data is important in the economic, finance and business sectors [11]. Forecasting
through the time series analysis technique was considered because the inflation rate is dynamic. Various famous time series techniques
available to forecast the data, such as autoregressive (AR), exponential smoothing (ES), moving average (MA), a combination of AR
and MA, known as the ARMA model and with the integration parameters involved, the autoregressive integrated moving average
(ARIMA) model [12,13]. Time series analysis, established by Box and Jenkin, has been used widely in research. Univariate discrete is
the starting point of time series analysis where a stochastic model describes how the series evolves. Based on this concept, forecasting is
performed in the serial data points and has several assumptions, such as the finite linear function is zt with independent noise of

* Corresponding author.
E-mail address: [email protected] (W.S. Alaloul).

https://doi.org/10.1016/j.heliyon.2024.e26037
Received 27 April 2023; Received in revised form 6 February 2024; Accepted 7 February 2024
Available online 8 February 2024
2405-8440/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
M.A. Musarat et al. Heliyon 10 (2024) e26037

random variables. In the autoregressive model, the assumption of zt is a continuous variance of the mean provided by the linear pattern
of the preceding z’s. This model can be considered for the stationary data series. In the case of non-stationary data, a differencing
parameter is introduced until the data become stationary. This observable fact is the Autoregressive Integrated Moving Average
(ARIMA) process, the ARIMA time series model [14,15].
ARIMA gained popularity because of its linear model characteristic for forecasting economic time series [16,17]. In addition to
ARIMA, artificial neural networks (ANNs) have also been used for their elastic and effective computational operations for the complex
relationships in the economic time series [18,19]. However, studies have shown a mixed overview on the choice of ARIMA and ANN;
therefore, both have been adopted in the forecasting techniques [20]. ARIMA model has shown remarkable performance in the ac­
curacy and precision of forecasting future values. The ARIMA model has been developed by introducing advanced machine learning
algorithms that refine the forecasting [12].
Another famous time series model is exponential smoothing (ES) [21]. ES is a simple yet effective forecasting tool [22] that
immediately forecasts discrete time series. This time series is popular due to its ease of forecasting and reliability [23]. ES helps
smoothen the original series whose moving average does comply with the time series analysis for future forecasting. In ES, consid­
eration is given to recent values in comparison to distant values [24,25]. ES is preferable on ARIMA for short-term forecasting due to
consideration of seasonality [26]. In ES, three basic models are preferred: trend-corrected ES [27], Simple ES [28] and Holt Winters’
model [29]. These models have distinctive features, including the assumption of time series on unobserved components (seasonality,
growth and level) and the adaptation of these components with the changes in the market [30]. Based on the previous progress of Ord,
Koehler [31], a statistical framework in ES was formulated by Hyndman, Koehler [26]. The framework functions with stochastic
models and allows computation of the smoothing parameters of likelihood estimates.
One of the reasons for resource price deviations leading to cost overrun is the inflation rate. Fluctuations in the inflation rate cause
prices to change annually, leaving a blackhole for stakeholders to fill until project completion [32]. Inflation in the economic world is
becoming inevitable, leaving adverse effects on industries and the economy [33,34]. The effects of the inflation rate have reached
beyond the construction industry and have also started to influence a country’s economy [35,36]. Experts have attempted to control
the inflation rate by introducing various policies; however, the issue remains complex because of the non-stagnant nature of money
[37–39]. Table 1 lists the countries where construction projects are affected by resource price deviation as influenced by the inflation
rate.
Generally, the reserve amount, known as the contingency cost, is kept in the project budget by the owner or related funding
agencies to deal with the unforeseen risks that increase real project costs [56,57]. The contingency cost burdens the project owner
because of the requirement to allot an extra amount to the actual project cost. Accounting for the effects of the cost overrun factors in
the initial budget estimation calculation is necessary for a project. Hence, developing an estimation model that can incorporate the
changing effects of the resources before submitting the Bill of Quantities is important [32]. Preparing a project budget is a challenge for
the contractor and the owner, and estimations should control for critical factors [3].
The volatile effect of the inflation rate has had adverse effects on the entire construction industry. Owners who avoid paying for
increased project costs leave contractors with no other option but to compromise on quality, which decreases project productivity.
Foreseeing the long-term changing effect of the inflation rate, which ultimately causes cost overrun in a construction project, has also
been a challenge for stakeholders. The inflation rate is one of the most critical factors of cost overrun in construction projects
worldwide, but its severity is still undetermined by stakeholders. The above-mentioned case clearly describes the need to incorporate
the inflation rate in construction rates during budget finalisation. However, current forecasting models cannot incorporate the precise
influencing criteria of the inflation rate on each construction rate. Therefore, in this study, a construction rate forecasting model was
developed by incorporating the impact of the inflation rate on the construction rates, i.e., building materials prices, labour wages and
machinery hire rates. Time series analysis was used because it is a popular forecasting technique. Artificial intelligence (AI) tools can
predict costs.
In contrast, the developed model adjusts the amount of influence that the inflation rate has on the construction rates. Managing an
inflation rate that deviates from the project budget is not an effortless task, and adjustments to the budget can be a challenge. This

Table 1
Cost overrun due to inflation in various countries.
S. No Country Cost Overrun

Material Labour Machinery

1 Malaysia [40,41] [42,43] [44]

2 Pakistan [45] – –
3 United Kingdom [46] – –
4 United States [47]
5 Egypt [48] – –
6 India [49] –
7 Afghanistan [50]
8 Uganda [51] – [51]
9 Nigeria [52] – –
10 Zambia [53]
11 Vietnam [54] –
12 Palestine [55] – –

2
M.A. Musarat et al. Heliyon 10 (2024) e26037

study provides a benchmark for construction industry stakeholders to adopt the construction rate forecasting model to avoid excessive
cost liability. No model can incorporate the inflation rate influence in estimating the project cost, raising the need for this model. The
main contributions of this study are as follows:

1. A mathematical construction rate forecasting model to deal with one of the major concerns of the construction industry, project cost
overrun, has been proposed.
2. The forecasted construction rates can be embedded into the Bill of Quantities before the tender allotment.
3. The construction industry stakeholders can forecast future rates in the present year and make the necessary adjustments to avoid a
project being cost overrun.

This study is organised to include the analysis and discussion of the construction rates forecasting outcome. First, the inflation rate
was predicted using Eviews software, which was incorporated into forecasting the construction rates based on the self-developed
model. Time series analysis was also used to forecast the construction rates to compare with the self-developed model based on
Akaike’s Information Criterion (AIC). This validation technique was utilised for building material prices, and the percentage deviation
validation technique for labour wages was optimised. No validation was performed for machinery hire rates because of fewer ob­
servations and the unavailability of the current year’s data.

Fig. 1. Research flowchart.

3
M.A. Musarat et al. Heliyon 10 (2024) e26037

2. Methodology

This study is an extensive version of the previous studies, where the relationship of construction rates, i.e., building materials prices
[41], labour wages [43] and machinery hire rates [44] were evaluated with the inflation rate. The data were collected from the
Government Departments of Malaysia, including the Construction Industry Development Board (CIDB) and the Department of Sta­
tistics Malaysia (DOSM) for 2013–2019. The relationship was evaluated using the Spearman correlation test with the value incor­
porated within the developed model for each construction rate. Thus, the thrust of the relationship can easily be evaluated. A
correlation parameter was introduced in the ES equation model by investigating the relationship and its impact. A detailed research
flow, which combines previous studies’ strategies, is presented in Fig. 1.

2.1. Forecasting model development

2.1.1. Self-developed forecasting model

A literature review shows that no forecasting model can incorporate fewer observations and no study has investigated the
behaviour of variables and ultimately forecast future values. Four mathematical equations that can forecast future rates were
developed. The modifications were brought by altering the existing ES model (Equation (1)), a type of time series model.
Exponential smoothing (ES) model
Ft+1 = α At + (1 – α) Ft , (1)

where F = forecast, A = Actual value (dependent variable), α = Smoothing Constant, t = time.

Based on ES, the modified equations (2)–(5) are as follows:
First equation
Ft+1 = α Yt + (1 – α) Ft + [r (Xt )] + ε, (2)
Second equation
Ft+1 = α Yt + (1 – α) Ft + [r (Xt )], (3)
Third equation
Ft+1 = α Yt + (1 – α) Ft + α (Xt ) + ε, (4)
Fourth equation
Ft+1 = α Yt + (1 – α) Ft + α (Xt ), (5)

where Y = dependent variable, X = independent variable, ε = linear error, r = correlation coefficient, α = 0.2, t = time, F = forecasting.
In the first equation, a linear error was introduced with the inclusion of correlation coefficient level and independent variable. Only
the correlation coefficient level and the independent variable were introduced in the second equation. In the third equation, the
correlation coefficient was replaced with the alpha with the inclusion of linear error. In the fourth equation, alpha was introduced
along with the independent variable.
The forecasting analysis of one variable (lorry) by ES equation and the four developed equations are discussed in Tables 2–6.
To access the best forecasting equation among the four, the mean square error (MSE) and mean absolute percentage error (MAPE)
were calculated using Equations (6) and (7) by putting a construction rate (lorry). The results are discussed in Table 7.

1∑ n 2
MSE = ̂ ,
(Yi − Yi) (6)
n i=1

Table 2
Forecasting analysis of ES equation.
Year Lorry Rates (Y) Forecasting Error Error 2

2013 12500 14895.86

2014 12,550 14416.69 − 1866.69 3,484,515.56
2015 12,600 14043.35 − 1443.35 2,083,255.10
2016 16900 13754.68 3145.32 9,893,045.09
2017 15444 14383.74 1060.26 1,124,144.72
2018 16573.67 14595.79 1977.88 3,911,991.62
2019 17703.33 14991.37 2711.96 7,354,729.35
2020 15533.76 15533.76 0.00 0.00
2021 15533.76 15533.76 0.00 0.00
2022 15533.76 15533.76 0.00 0.00
2023 15533.76 15533.76 0.00 0.00
2024 15533.76 15533.76 0.00 0.00
2025 15533.76

4
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 3
Forecasting analysis of the first equation.
Year Inflation Rate (X) Lorry Rates (Y) Predicted (Y) Residual Forecasting Error Error 2

2013 2.11 12500 14965.68 − 2465.68 14895.86

2014 3.14 12,550 13912.09 − 1362.09 11949.65 600 360,422.54
2015 2.1 12,600 14975.91 − 2375.91 10705.61 1894 3,588,731.02
2016 2.08 16900 14996.37 1903.63 8707.22 8193 67,121,578.69
2017 3.8 15444 13236.98 2207.02 12248.07 3196 10,213,944.56
2018 1 16573.67 16101.10 472.57 15091.84 1482 2,195,833.18
2019 1.02 17703.33 16080.64 1622.69 15860.13 1843 3,397,388.14
2020 2.67 17850.80 14396.60 3454.21 17850.80 0.00 0.00
2021 1.91 21303.29 15165.41 6137.88 21303.29 0.00 0.00
2022 1.64 27439.94 15446.06 11993.88 27439.94 0.00 0.00
2023 1.89 39432.76 15194.19 24238.57 39432.76 0.00 0.00
2024 2.51 63670.12 14557.87 49112.25 63670.12 0.00 0.00
2025 112780.76 112780.76 0.00 0.00

Table 4
Forecasting analysis of the second equation.
Year Inflation Rate (X) Lorry Rates (Y) Forecasting Error Error 2

2013 2.11 12500 14895.86

2014 3.14 12,550 14415.33 − 1865 3,479,452.22
2015 2.1 12,600 14040.24 − 1440 2,074,303.26
2016 2.08 16900 13750.85 3149 9,917,177.00
2017 3.8 15444 14379.34 1065 1,133,503.92
2018 1 16573.67 14589.83 1984 3,935,631.18
2019 1.02 17703.33 14985.95 2717 7,384,137.89
2020 2.67 15528.77 15528.77 0.00 0.00
2021 1.91 15527.06 15527.06 0.00 0.00
2022 1.64 15525.83 15525.83 0.00 0.00
2023 1.89 15524.77 15524.77 0.00 0.00
2024 2.51 15523.56 15523.56 0.00 0.00
2025 15521.95 15521.95

Table 5
Forecasting analysis of the third equation.
Year Inflation Rate (X) Lorry Rates (Y) Predicted (Y) Residual Forecasting Error Error 2

2013 2.11 12500 14965.681 − 2465.681 14895.86

2014 3.14 12,550 13912.094 − 1362.094 11951.43 599 358,289.98
2015 2.1 12,600 14975.91 − 2375.91 10709.68 1890 3,573,327.20
2016 2.08 16900 14996.368 1903.632 8712.25 8188 67,039,245.20
2017 3.8 15444 13236.98 2207.02 12253.85 3190 10,177,068.27
2018 1 16573.67 16101.1 472.57 15099.66 1474 2,172,709.64
2019 1.02 17703.33 16080.642 1622.688 15867.23 1836 3,371,260.01
2020 2.67 17857.34 14396.59517 3460.747526 17857.34 0.00 0.00
2021 1.91 21318.62 15165.41128 6153.212214 21318.62 0.00 0.00
2022 1.64 27472.22 15446.06426 12026.15574 27472.22 0.00 0.00
2023 1.89 39498.70 15194.18918 24304.51082 39498.70 0.00 0.00
2024 2.51 63803.59 14557.86653 49245.72347 63803.59 0.00 0.00
2025 113049.82 113049.82

where n = number of data points, Yi = observed values, Y

̂ i = predicted values.
⃒ ⃒
1∑ n ⃒ ⃒
⃒ At − Ft ⃒ ,
MAPE = (7)
n t=1 ⃒ At ⃒

where n = number of times the summation iteration happens, At = actual value, Ft = forecast value.
Based on the results, the lowest values for MSE and MAPE were from the fourth and second equations. However, the fourth equation
was rejected because it did not incorporate the level of relationship among the variables. The second equation was chosen for further
assessment as the correlation coefficient (r) is involved. As discussed, correlation relationships have five levels. The second equation
can incorporate the correlation level, and the future rates can be predicted based on the results. By adopting the second equation, the
existing deficiency of the ES model can be eliminated which does not incorporate autocorrelations.

5
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 6
Forecasting analysis of the fourth equation.
Year Inflation Rate (X) Lorry Rates (Y) Forecasting Error Error 2

2013 2.11 12500 14895.86

2014 3.14 12,550 14417.11 − 1867 3,486,091.22
2015 2.1 12,600 14044.31 − 1444 2,086,043.43
2016 2.08 16900 13755.87 3144 9,885,545.05
2017 3.8 15444 14385.11 1059 1,121,241.53
2018 1 16573.67 14597.65 1976 3,904,653.24
2019 1.02 17703.33 14993.05 2710 7,345,594.02
2020 2.67 15535.31 15535.31 0.00 0.00
2021 1.91 15535.85 15535.85 0.00 0.00
2022 1.64 15536.23 15536.23 0.00 0.00
2023 1.89 15536.56 15536.56 0.00 0.00
2024 2.51 15536.94 15536.94 0.00 0.00
2025 15537.44 15537.44

Table 7
MSE and MAPE.
Forecasting Equations First Equation Second Equation Third Equation Fourth Equation

MSE 14479650 4654034 14448650 4638195

MAPE 10% 7.19% 10% 7.18%

2.1.2. Time series forecasting model

Comparisons with existing forecasting models are necessary to check the accuracy of the self-developed forecasting model. In this
study, three construction rates were taken: building materials prices, labour wages and machinery hire rates, but only building ma­
terials prices data was available monthly. Because the existing time series model cannot make predictions based on a small number of
observations, therefore, the forecasting of building materials prices data was performed based on the autoregressive integrated moving
average (ARIMA) time series model. Once the forecasting of the monthly data of building materials prices was completed, it was
converted into annual data for comparison with the data generated by the self-developed forecasting model. Before performing any
time series analysis, four data components need to be examined: trend, seasonality, cyclical and irregular. The trend can be upward or
downward in a linear or nonlinear pattern. Seasonality is a short-term regular pattern that can be over a year or months. Cyclical also
has a regular pattern, but its pattern is in the long term. Irregular is unpredictable because of frequent variations in the data by an
influential factor.

2.1.2.1. ARIMA model. ARIMA is one of the most popular techniques in time series analysis for forecasting. It is the combination of
three major components: autoregressive (AR), integrated (I) and moving average (MA). AR and MA individually can also forecast the
data; however, their accuracy is not absolute. The autoregressive moving average model has also been utilised, but introducing the
integrated variable makes the results more promising. Thus, the ARIMA model was used to forecast the monthly building materials
price data in this study. Performing ARIMA can be achieved in two ways: one is a traditional programming method, and the other is
utilising the built-in software. The number of variables was high; thus, software was chosen for the analysis.

2.1.2.2. EViews software. EViews is a statistical package utilised for econometric analysis, mainly incorporated with time series. For
the study’s purpose, EViews Student Version Lite (V-11) was used for the analysis. EViews has a built-in option to perform ARIMA time
series analysis where the p, d and q values are selected automatically based on the best combination of the model. The AR value is
represented by p, the I value is represented by d, known as differencing, and the MA value is represented by q. Moreover, in SAR and
SMA, S represents the seasonality in the data.

2.1.3. Model validation

Model validation in conducting analysis using two models. This study used Akaike’s Information Criterion (AIC) to validate both
models. The lower the AIC value, the better the forecasting model. The AIC, using Equation (8) can be calculated as follows:
AIC = − 2 ln(L) + 2K (8)

where L = maximum value of the likelihood function for the model and K = number of independent values. By default, the K value is 2,
and the addition is based on the number of independent variables. In this case, the number of independent variables was 1, i.e., the
inflation rate, and therefore, the K value was taken as 3.
Building materials prices model validation, labour wages and machinery hire rates were also validated by percentage deviation
calculation.

6
M.A. Musarat et al. Heliyon 10 (2024) e26037

3. Results and discussion

3.1. Construction rates forecasting

The self-developed forecasting model (second equation) was applied to all construction rates. The forecasting of the construction
rates was made from 2020 to 2025.

3.1.1. Inflation rate forecasting

The independent variable should be forecasted first to enable the self-developed forecasting model to work. The EViews software
was used to forecast the inflation rate, as shown in Fig. 2. The inflation rate data were taken from 1980 to 2019 for forecasting
purposes. Data from 1980 to 2010 were considered as train, and data from 2011 to 2019 were considered as a test. To forecast the
inflation rate, 25 ARIMA models were simulated, as shown in Fig. 3, where ARMA model (2,2)(0,0) was the most suitable model based
on the lowest AIC value, as shown in Table 8. When selecting the ARIMA model parameters, the maximum differencing was kept as 2,
and the maximum AR and MA values as 4.

3.1.2. Building materials prices forecasting

The building materials prices were forecasted using the second equation and incorporating the forecasted inflation rate, as shown in
Table 9. The table shows that most materials types displayed an increasing trend, indicating that material prices will go higher.
The first forecasted value for 2013 was computed by taking the average value of the dependent variable (building material) because
the required F was not available at the initial stage. In this case, the literature supports taking the average or initial value as the first F
value. Once the end of 2019 is reached in the dependent variable, an average value of the forecasted values from 2013 to 2019 was
taken to forecast the later years. The role of the correlation coefficient is significant because it indicates the thrust of the relationship.
Positive and negative relationships influence the forecasted rates.
Moreover, if the relationship is strong, the inclination or declination behaviour of the forecasting will also be high. For the materials
where the relationship of the inflation rate with materials prices is weak or moderate, the inclination or declination behaviour of the
forecasting value will also be weak or moderate. A comparison was made with the existing ARIMA model, a time series model, via
EViews to prove the effectiveness of the self-developed model.

3.1.2.1. ARIMA forecasting for building materials prices. A high number of observations are required to perform the ARIMA forecasting.
However, the monthly data on the building materials prices and the inflation rate were also available, making it easier to conduct the
analysis. The monthly inflation rate was initially forecasted via EViews by performing the Automated Arima Forecasting function. The
output is shown in Fig. 4.
The inflation rate values were taken as the independent variable. Later forecasting was conducted for the individual building
materials. The building materials prices data were taken from January 2013 to December 2019 for forecasting purposes, where data
from January 2013 to June 2018 were considered as train and data from July 2018 to December 2019 were considered as a test. The
maximum differencing value was taken as 2, AR as 4, MA as 4, SAR as 2 and SMA as 2. Various forecasting models were generated, and
the best one was chosen for each material, as shown in Table 10.

3.1.2.2. Validation of building materials models. To validate the self-developed and ARIMA models of each building material price, AIC
was calculated and presented in Table 11.

Fig. 2. Inflation rate forecasting.

7
M.A. Musarat et al. Heliyon 10 (2024) e26037

Fig. 3. ARIMA model comparison.

Table 8
Forecasting model AIC values.
S. No Model AIC

1 (2,2) (0,0) 3.929456

2 (3,2) (0,0) 3.985367
3 (2,3) (0,0) 3.985724
4 (0,3) (0,0) 4.033902
5 (4,0) (0,0) 4.043242
6 (3,3) (0,0) 4.051557
7 (4,2) (0,0) 4.054188
8 (1,0) (0,0) 4.061613
9 (1,3) (0,0) 4.068897
10 (0,4) (0,0) 4.072887
11 (4.1) (0,0) 4.075969
12 (1,2) (0,0) 4.076764
13 (2,4) (0,0) 4.09399
14 (3,1) (0,0) 4.102562
15 (2,0) (0,0) 4.117855
16 (1,1) (0,0) 4.121832
17 (4,4) (0,0) 4.128053
18 (1,4) (0,0) 4.133878
19 (3,0) (0,0) 4.147293
20 (0,1) (0,0) 4.151659
21 (0,2) (0,0) 4.151845
22 (2,1) (0,0) 4.178257
23 (4,3) (0,0) 4.341339
24 (0,0) (0,0) 4.351245
25 (3,4) (0,0) 4.400444

Table 11 shows that all the AIC values of the self-developed model are lesser than the ARIMA model, proving that the model yields
the best alternative for forecasting. The main benefit of this model is that it provides forecasting even for a smaller number of ob­
servations. Most importantly, the self-developed model incorporates the relationship behaviour of the independent and dependent
variables based on the correlation, which has a major impact.

3.1.3. Labour wages forecasting

Labour wages were also computed from the self-developed model. The forecasting results are shown in Table 12. The model adopts
the same strategy of calculating future wages as described above. Initially, the correlation coefficients of all the wages required to
forecast the values from 2020 to 2025 were computed.

8
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 9
Forecasted building materials prices.
Building Materials Prices Forecasted Rates (RM)

S. No Category of Material 2020 2021 2022 2023 2024 2025

1.1 Ordinary Portland Cement, 50 kg bag 21.74 23.70 24.73 25.42 26.22 27.43
1.2 Ordinary Portland Cement in Bulk 299.18 306.19 310.77 314.00 316.70 319.37
2.1 Granite Aggregate 3/4″ 37.93 37.98 38.00 38.01 38.04 38.08
3.1 Normal River Sand- Ex 37.79 36.92 36.41 36.07 35.72 35.27
3.2 Fine River Sand for Plastering -Ex 39.82 39.15 38.75 38.48 38.22 37.88
3.3 Normal Mining Sand 34.72 34.23 33.95 33.75 33.56 33.30
3.4 Fine Mining Sand for Plastering 37.56 36.40 35.75 35.32 34.85 34.20
4.1 Mild Steel Round Bar R10, MS146 2410.25 2404.00 2399.50 2396.26 2393.94 2392.32
4.2 Mild Steel Round Bar R12, MS146 2444.99 2437.40 2431.95 2428.02 2425.20 2423.22
4.3 Mild Steel Round Bar R16, MS146 2402.20 2389.75 2380.99 2374.71 2370.05 2366.47
4.4 High Tensile Deformed Bar-Y10, MS146 2326.20 2318.75 2313.44 2309.61 2306.83 2304.83
4.5 High Tensile Deformed Bar-Y12, MS146 2304.92 2296.93 2291.23 2287.12 2284.15 2282.00
4.6 High Tensile Deformed Bar-Y16, MS146 2195.16 2190.00 2186.26 2183.55 2181.65 2180.39
4.7 High Tensile Deformed Bar-Y20, MS146 2176.05 2169.31 2164.44 2160.91 2158.42 2156.73
4.8 High Tensile Deformed Bar-Y25, MS146 2176.05 2169.31 2164.44 2160.91 2158.42 2156.73
4.9 High Tensile Deformed Bar-Y32, MS146 2176.05 2169.31 2164.44 2160.91 2158.42 2156.73
4.10 BRC A6, MS145 61.56 59.10 57.69 56.73 55.74 54.43
4.11 BRC A7, MS145 83.10 81.32 80.25 79.52 78.82 77.97
4.12 BRC A8, MS145 109.85 107.58 106.21 105.27 104.37 103.30
4.13 BRC A9, MS145 137.39 134.74 133.11 131.98 130.94 129.73
4.14 BRC A10, MS145 170.18 166.94 164.90 163.47 162.21 160.84
5.1 Ready Mix Concrete - Normal Mix - Grade15, Granite 183.64 186.19 187.71 188.75 189.76 191.00
5.2 Ready Mix Concrete - Normal Mix - Grade 20, Granite 190.31 193.08 194.74 195.88 196.97 198.31
5.3 Ready Mix Concrete - Normal Mix - Grade 25, Granite 201.38 204.00 205.57 206.65 207.69 208.95
5.4 Ready Mix Concrete - Normal Mix - Grade 30, Granite 212.45 215.11 216.71 217.80 218.85 220.12
5.5 Ready Mix Concrete - Normal Mix - Grade 35, Granite 222.72 225.30 226.84 227.89 228.91 230.16
6.1 Common Clay Bricks - Pallet 34.91 33.40 32.61 32.09 31.47 30.52
6.2 Cement Sand Bricks - Pallet 24.60 22.95 22.12 21.57 20.89 19.82
7.1 Interlocking Concrete Tiles - Standard Duotone Colour 420 mm × 330 mm 90.79 90.25 89.98 89.81 89.59 89.23
7.3 MS Decking - Ajiya AP Rib Hi-Tensile G26, 0.47 mm TCT, Clean Colourbond 46.50 44.33 43.16 42.39 41.50 40.21
(Commercial)
7.4 MS Decking - Ajiya AP Rib Hi- Tensile G24, 0.53 mm TCT, Clean Colourbond 55.95 53.94 52.87 52.16 51.33 50.12
(Commercial)
7.5 MS Decking - Ajiya Euro Step Roofing M350 G28, 0.40 mm TCT, Clean 40.07 37.96 36.84 36.10 35.23 33.96
Colourbond
7.6 MS Decking - Ajiya Euro Step Roofing M350 G26, 0.47 mm TCT, Clean 52.88 50.54 49.20 48.30 47.36 46.12
Colourbond
7.11 Corrugated Roofing Sheet-76mm Double Width, 1065 mm × 2440 mm x 4 mm 22.85 20.81 19.72 19.01 18.17 16.92
(Hume/Malex/UAC)
8.4 Plain Homogeneous Floor Tiles, Standard light Colour, 300 mm × 300 mm x 8 52.87 52.03 51.60 51.31 50.97 50.43
mm - Grade 1
8.6 Glazed Ceramic Wall Tiles, Standard light Colour, 200 mm × 250 mm x 6 mm - 68.34 66.79 65.94 65.37 64.74 63.86
Grade 1
9.1 Plain Cellulose Fibre Ceiling Sheet,610 mm × 1220 mm x3.2 mm (UAC/ 40.28 39.75 39.51 39.36 39.13 38.75
MALEX/HUME)
9.2 Plain Cellulose Fibre Ceiling Sheet,610 mm × 1220 mm x 4.5 mm (UAC/ 69.12 67.26 66.27 65.61 64.85 63.74
MALEX/HUME)
9.3 Plain Gypsum Board, 610 mm × 1220 mm x 9.5 mm, (Boral/Armstrong) 54.82 52.67 51.46 50.65 49.78 48.59
10.4 PVC Pressure Pipes Class D Grey Colour - 50 mm Diameter (paling), MS628 33.76 32.65 31.96 31.47 31.04 30.55
10.5 PVC Pressure Pipes Class E Grey Colour - 25 mm Diameter (paling), MS628 13.29 12.47 11.99 11.66 11.33 10.93
10.7 HDPE Pipe PN 16–25 mm Diameter, MS 1058 141.33 139.16 137.90 137.05 136.18 135.08
10.8 HDPE Pipe PN 16–50 mm Diameter, MS 1058 469.54 465.09 462.12 460.02 458.32 456.74
11.1 Polyethene water tank (Polytank), Circular- 200 gallons, MS 1225-weida 193.00 183.74 177.44 172.95 169.44 166.40
equivalent
11.5 Water Closet Western Type- WC 644, white colour without cistern, claytan 184.34 180.68 178.24 176.51 175.11 173.81
11.7 Urinal Bowl, santana 320 c/w hanger, flange, ceramic waste & cleaning set, 504.50 497.14 492.17 488.64 485.84 483.35
Johnson-Suisse
11.8 Stainless Steel Sinks - Single Bowl Single Drainer- Lay On Type, 42″ × 18″ 96.36 93.77 92.25 91.22 90.19 88.88
12.1 Paint-ICI dulux standard colour-Undercoat speed 100.56 102.26 103.17 103.76 104.46 105.50
12.2 Paint-ICI dulux standard colour-external acrylic emulsion, weathershield 193.37 184.27 178.16 173.82 170.35 167.20
12.7 Paint-Nippon standard colour-timber/wood, timber finish 108.59 105.72 103.94 102.71 101.59 100.32
12.12 Paint - ICI Dulux Standard Colour Gloss Oil-based, Gloss Finish (5L) 117.59 115.17 113.85 112.98 111.99 110.57
13.1 Square hollow sections-12mm × 12 mm x 1.0 mm (0.357 kg/m) 3164.91 3149.87 3139.46 3132.01 3126.35 3121.73
13.2 Square Hollow Sections - 50 mm × 50 mm x 3.0 mm (4.38 kg/m) 3232.01 3225.95 3221.72 3218.69 3216.42 3214.61
13.3 Square Hollow Sections - 150 mm × 150 mm x 4.0 mm (20.20 kg/m) 3309.30 3311.54 3313.00 3314.02 3314.88 3315.73
13.4 Universal beams-102mm × 102 mm x 8.76 mm (19.35 kg/m) 2941.63 2964.46 2980.47 2991.96 3000.51 3007.17
13.5 Universal beams-400mm × 400 mm (2.17 kg/m) 3152.22 3186.83 3211.24 3228.78 3241.71 3251.52
(continued on next page)

9
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 9 (continued )
Building Materials Prices Forecasted Rates (RM)

S. No Category of Material 2020 2021 2022 2023 2024 2025

13.6 Equal angles-38mm × 38 mm x 3.8 mm (2.17 kg/m) 2636.51 2644.35 2649.73 2653.57 2656.53 2659.04
13.7 Equal angles-50mm × 50 mm x 4 mm (3.06 kg/m) 2559.78 2576.86 2588.82 2597.39 2603.79 2608.82
14.1 Plywood-shuttering board, 4 x 8 × 12 mm 48.65 47.15 46.37 45.87 45.25 44.29
15.1 GMS Heavy Hardwood, Balau 1 4942.11 4931.14 4923.39 4917.82 4913.72 4910.62
15.2 GMS light Hardwood, Dark Red Meranti 3181.03 3175.37 3171.37 3168.50 3166.38 3164.79
15.3 GMS Medium Hardwood, Kapur 3291.38 3279.85 3271.69 3265.82 3261.51 3258.29
15.4 Scantling Medium Hardwood, Kapur 3149.67 3137.38 3128.67 3122.42 3117.83 3114.40
16.1 Clear Float Glass 5 mm Thk, Local/Imported 54.38 52.21 50.99 50.17 49.29 48.09
16.3 Tinted Float Glass 6 mm Thk, Local/Imported, Inclusive of Cutting 66.76 64.62 63.41 62.61 61.74 60.55
17.1 CL03-Duraset brand residential grade 3 cylindric knob sets - PRIVACY- 45.24 46.68 47.54 48.13 48.70 49.40
CL0363010
17.2 CL03-Duraset brand residential grade 3 cylindric knob sets - PATIO- 48.76 50.47 51.45 52.12 52.80 53.69
CL0363020
17.3 CL03-Duraset brand residential grade 3 cylindric knob sets - PASSAGE- 44.15 43.57 43.29 43.12 42.87 42.45
CL0363030
17.4 Concorde 102 mm × 76 mm x 2.0 mm SS hinge SUS 304 24.46 23.66 23.25 22.99 22.66 22.14
17.5 CL03- Duraset Brand Residential Grade 3 Cylindric Knob sets - ENTRANCE - 49.96 52.07 53.31 54.16 55.00 56.06
CL0363000

Fig. 4. Inflation rate forecasting (month-wise).

Table 10
ARIMA model for individual materials.
S. No ARIMA Model S. No ARIMA Model S. No ARIMA Model S. No ARIMA Model

1.1 (1,0) (0,0) 1.2 (1,0) (0,0) 2.1 (2,1) (0,0) 3.1 (1,0) (0,0)
3.2 (1,0) (0,0) 3.3 (2,1) (0,0) 3.4 (1,0) (0,0) 4.1 (1,1) (0,0)
4.2 (1,1) (0,0) 4.3 (2,1) (0,0) 4.4 (1,1) (0,0) 4.5 (1,1) (0,0)
4.6 (1,1) (0,0) 4.7 (1,1) (0,0) 4.8 (1,1) (0,0) 4.9 (1,1) (0,0)
4.10 (2,2) (0,0) 4.11 (2,2) (0,0) 4.12 (2,2) (0,0) 4.13 (1,2) (0,0)
4.14 (0,0) (0,0) 5.1 (2,1) (0,0) 5.2 (1,0) (0,0) 5.3 (1,0) (0,0)
5.4 (2,1) (0,0) 5.5 (1,0) (0,0) 6.1 (1,0) (1,1) 6.2 (1,0) (0,0)
7.1 (2,1) (1,1) 7.3 (0,0) (0,0) 7.4 (0,0) (0,0) 7.5 (2,2) (0,0)
7.6 (0,0) (0,0) 7.11 (0,1) (0,0) 8.4 (1,1) (0,0) 8.6 (1,0) (0,0)
9.1 (3,2) (1,1) 9.2 (2,2) (0,0) 9.3 (0,0) (0,0) 10.4 (2,3) (0,0)
10.5 (0,3) (0,0) 10.7 (0,0) (0,0) 10.8 (0,0) (1,1) 11.1 (0,0) (0,0)
11.5 (0,1) (0,0) 11.7 (0,0) (0,0) 11.8 (0,0) (1,1) 12.1 (4,3) (0,1)
12.2 (0,0) (0,0) 12.7 (2,2) (0,0) 12.12 (2,1) (0,0) 13.1 (0,0) (0,0)
13.2 (0,3) (0,0) 13.3 (0,3) (0,0) 13.4 (0,1) (0,0) 13.5 (0,0) (0,0)
13.6 (1,0) (0,0) 13.7 (1,0) (0,0) 14.1 (2,1) (0,0) 15.1 (0,0) (0,0)
15.2 (0,0) (0,0) 15.3 (4,4) (1,0) 15.4 (2,1) (0,0) 16.1 (0,0) (0,0)
16.3 (0,0) (0,0) 17.1 (1,0) (0,0) 17.2 (1,0) (0,0) 17.3 (1,3) (0,0)
17.4 (0,0) (0,0) 17.5 (1,0) (1,1)

10
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 11
AIC validation.
S. No ARIMA AIC Self-Model AIC S. No ARIMA AIC Self-Model AIC S. No ARIMA AIC Self-Model AIC

1.1 1.734 − 0.882 1.2 8.379 − 9.311 2.1 4.788 0.034

3.1 5.168 − 1.430 3.2 5.068 − 1.014 3.3 5.256 − 0.800
3.4 5.116 − 1.175 4.1 12.599 − 13.967 4.2 12.361 − 14.048
4.3 12.639 − 14.155 4.4 12.900 − 13.679 4.5 12.901 − 14.137
4.6 12.798 − 13.461 4.7 12.798 − 13.931 4.8 12.798 − 13.931
4.9 12.798 − 13.931 4.10 5.015 − 3.055 4.11 5.475 − 2.571
4.12 5.929 − 3.926 4.13 6.269 0.422 4.14 − 3.322 0.133
5.1 6.663 − 5.453 5.2 6.679 − 5.598 5.3 6.885 − 5.862
5.4 7.079 − 6.119 5.5 6.904 − 5.805 6.1 3.344 0.101
6.2 2.936 0.152 7.1 2.510 − 1.907 7.3 3.643 − 1.255
7.4 − 5.086 − 0.941 7.5 − 4.768 − 1.212 7.6 2.494 − 2.427
7.11 − 4.832 − 0.829 8.4 − 4.526 0.795 8.6 5.225 − 1.999
9.1 − 4.404 1.113 9.2 − 5.005 − 1.030 9.3 3.426 − 1.901
10.4 − 4.346 − 1.450 10.5 0.450 0.848 10.7 4.664 − 2.739
10.8 6.546 − 7.774 11.1 − 2.449 − 10.764 11.5 5.611 − 6.634
11.7 7.191 − 9.538 11.8 3.983 − 3.221 12.1 4.618 − 1.532
12.2 − 1.538 − 10.679 12.7 4.782 − 4.485 12.12 − 4.730 − 2.088
13.1 11.226 − 13.254 13.2 10.599 − 11.448 13.3 10.106 − 10.317
13.4 10.508 − 14.476 13.5 11.742 − 16.589 13.6 10.909 − 13.325
13.7 11.298 − 14.831 14.1 3.701 − 1.021 15.1 13.086 − 14.897
15.2 11.008 − 12.388 15.3 − 5.035 − 14.992 15.4 11.913 − 15.277
16.1 3.174 − 2.349 16.3 4.031 − 1.873 17.1 4.521 − 1.520
17.2 4.726 − 1.992 17.3 4.335 2.276 17.4 − 2.862 2.430
17.5 5.935 − 2.968

3.1.3.1. Validation of labour wages model. The fewer observations for each labour wages category makes it impossible to compare
labour wages forecasted values with any other forecasting model. Therefore, the forecasted labour wages with the actual labour wages
data published in 2020 were compared, and the percentage deviation was calculated. The results are shown in Table 13.
Table 13 shows the overall deviation between the forecasted and actual labour wages in 2020. An overall deviation of 5.69%
occurred, indicating that the forecasted wages are near the published wages. The effects of the COVIID-19 pandemic were not
considered, which could impact the actual rates in 2020.

3.1.4. Machinery hire rates forecasting

Machinery hire rates were also computed from the self-developed model, and the forecasting results are shown in Table 14. The
model adopts the same strategy of calculating future wages as described above. Initially, the correlation coefficients of all machinery
hire rates, which were required to forecast the values from 2020 to 2025, were computed.
In the case of machinery hire rates, the forecasted rates with the actual rates could not be validated due to the data’s unavailability.
However, by looking at the upward scenario, it can be concluded that the model can forecast the rates near the existing rates. Field
verification by a semi-structured interview was also performed. Most of the stakeholders with various industry working experiences
agreed with the forecasting model’s formulation and endorsed its workability for real-time construction projects. The demographic
profile of the experts is shown in Table 15.

4. Discussion

Cost overrun in the construction industry has existed for decades, and the issue persists even with various reforms. One of the major
reasons for its persistence is the inflation rate, which completely deviates from the construction rates. However, most project budget
preparations do not incorporate the inflation rate. Therefore, this study proposed a forecasting model that can understand the changing
behaviour of the inflation rate over time. Another issue aside from the absence of a construction rate forecasting model is that all pre-
existing forecasting model requires a high number of observations to make forecasts. However, a self-developed model can perform
forecasting even with fewer observations. The trend of the forecasted values is highly dependent on the relationship between the
inflation rate and construction rates. The workability of the self-developed model can be achieved during the project budget prepa­
ration and development of the Bill of Quantities. By putting the construction and inflation rates in the model, a forecasted value for any
particular year can be taken, which can then be incorporated into the budget estimation. The model can yield better results for projects
with long duration because those projects have longer exposure to the changing behaviour of the inflation rate. Although the analysis
was conducted on Malaysia’s construction rates, the model applies to the construction industry of any country where the inflation rate
deviates from the rates and causes cost overrun.

5. Limitations of study

This study has several limitations:

11
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 12
Forecasted labour wages.
Labour Wages Forecasted Rates (RM)

S. No Category of Labour 2020 2021 2022 2023 2024 2025

Construction Workers Group

10131 General Construction Worker - Building, Local (Helper) 64.50 63.06 62.26 61.73 61.14 60.32
10132 General Construction Worker - Building, Foreign (Helper) 53.05 52.10 51.55 51.18 50.79 50.29
10211 Concretor, Skilled, Local 90.59 90.60 90.63 90.65 90.65 90.63
10212 Concretor, Skilled, Foreign 72.19 70.91 70.18 69.69 69.18 68.50
10221 Concretor, Semi-skilled, Local 68.08 66.50 65.61 65.01 64.37 63.51
10222 Concretor, Semi-skilled, Foreign 58.48 57.15 56.40 55.89 55.36 54.64
10311 Bricklayer, Skilled, Local 98.95 99.07 99.15 99.20 99.25 99.28
10312 Bricklayer, Skilled, Foreign 76.58 75.80 75.38 75.09 74.78 74.32
10321 Bricklayer, Semi-skilled, Local 71.05 69.30 68.35 67.72 67.01 65.99
10322 Bricklayer, Semi-skilled, Foreign 58.66 56.80 55.80 55.13 54.37 53.29
10411 Plasterer, Skilled, Local 111.13 111.97 112.44 112.75 113.10 113.57
10412 Plasterer, Skilled, Foreign 84.38 83.18 82.52 82.08 81.60 80.92
10421 Plasterer, Semi-skilled, Local 76.80 75.03 74.07 73.43 72.71 71.68
10422 Plasterer, Semi-skilled, Foreign 64.24 63.06 62.39 61.95 61.47 60.81
10511 Tiler, Skilled, Local 121.98 122.12 122.16 122.17 122.24 122.40
10512 Tiler, Skilled, Foreign 94.65 94.91 95.00 95.05 95.17 95.42
10521 Tiler, Semi-skilled, Local 90.13 88.93 88.27 87.83 87.34 86.65
10522 Tiler, Semi-skilled, Foreign 70.96 69.63 68.87 68.36 67.82 67.09
10611 Barbender, Skilled, Local 108.06 108.94 109.39 109.68 110.05 110.62
10612 Barbender, Skilled, Foreign 83.11 83.28 83.37 83.43 83.50 83.61
10621 Barbender, Semi-skilled, Local 80.63 79.89 79.43 79.11 78.82 78.50
10622 Barbender, Semi-skilled, Foreign 63.39 62.75 62.39 62.14 61.88 61.53
10711 Carpenter - Formwork, Skilled, Local 104.51 105.07 105.44 105.70 105.92 106.13
10712 Carpenter - Formwork, Skilled, Foreign 79.59 78.52 77.96 77.59 77.16 76.50
10721 Carpenter - Formwork, Semi-skilled, Local 77.53 76.61 76.12 75.80 75.42 74.87
10722 Carpenter - Formwork, Semi-skilled, Foreign 63.58 63.86 64.04 64.17 64.27 64.38
10811 Carpenter - Joinery, Skilled, Local 126.45 128.30 129.33 130.01 130.77 131.82
10812 Carpenter - Joinery, Skilled, Foreign 94.02 94.46 94.68 94.82 95.00 95.30
10821 Carpenter - Joinery, Semi-skilled, Local 83.86 83.18 82.82 82.59 82.31 81.88
10822 Carpenter - Joinery, Semi-skilled, Foreign 65.49 63.59 62.55 61.86 61.08 59.98
10911 Roofer, Skilled, Local 108.12 108.94 109.43 109.76 110.09 110.50
10912 Roofer, Skilled, Foreign 84.85 84.85 84.82 84.79 84.79 84.85
10921 Roofer, Semi-skilled, Local 79.76 78.81 78.31 77.97 77.58 77.02
10922 Roofer, Semi-skilled, Foreign 60.88 58.68 57.47 56.66 55.77 54.50
11011 Steel Structure Fabricator, Skilled, Local 118.59 119.36 119.86 120.21 120.51 120.81
11012 Steel Structure Fabricator, Skilled, Foreign 92.83 93.56 93.94 94.19 94.49 94.93
11021 Steel Structure Fabricator, Semi-skilled, Local 87.00 87.11 87.20 87.28 87.31 87.31
11022 Steel Structure Fabricator, Semi-skilled, Foreign 69.37 69.11 68.96 68.86 68.76 68.62
11111 General Welder, Skilled, Local 125.28 127.54 128.90 129.84 130.73 131.80
11112 General Welder, Skilled, Foreign 97.52 99.12 100.04 100.67 101.31 102.15
11121 General Welder, Semi-skilled, Local 87.64 87.59 87.64 87.69 87.66 87.49
11122 General Welder, Semi-skilled, Foreign 72.04 72.10 72.16 72.21 72.23 72.21
11211 Plumber - Building & Sanitary, Skilled, Local 114.01 114.66 115.11 115.43 115.67 115.88
11212 Plumber - Building & Sanitary, Skilled, Foreign 90.51 90.23 90.06 89.95 89.84 89.70
11221 Plumber - Building & Sanitary, Semi-skilled, Local 82.61 81.64 81.13 80.79 80.39 79.79
11222 Plumber - Building & Sanitary, Semi-skilled, Foreign 64.91 62.67 61.45 60.64 59.73 58.42
11311 Plumber - Reticulation, Skilled, Local 113.78 113.87 113.96 114.02 114.05 114.05
11312 Plumber - Reticulation, Skilled, Foreign 89.04 88.86 88.73 88.64 88.58 88.53
11321 Plumber - Reticulation, Semi-skilled, Local 82.88 81.95 81.47 81.15 80.77 80.18
11322 Plumber - Reticulation, Semi-skilled, Foreign 69.86 69.17 68.79 68.54 68.26 67.86
11421 Building Wiring Installer, Semi-skilled, Local 98.62 98.91 99.11 99.26 99.37 99.45
11422 Building Wiring Installer, Semi-skilled, Foreign 79.40 79.55 79.65 79.72 79.78 79.83
11511 Electrical Wireman PW2, Skilled, Local (per day) 135.46 137.89 139.34 140.34 141.30 142.48
11512 Electrical Wireman PW2, Skilled, Foreign (per day) 105.80 107.32 108.20 108.80 109.40 110.18
11611 Electrical Wireman PW4, Skilled, Local (per day) 159.48 160.72 161.51 162.06 162.54 163.06
11612 Electrical Wireman PW4, Skilled, Foreign (per day) 121.84 122.36 122.67 122.88 123.09 123.35
11711 Scaffolder - Prefabricated, Skilled, Local 99.75 100.32 100.69 100.95 101.16 101.38
11712 Scaffolder - Prefabricated, Skilled, Foreign 82.03 82.07 82.10 82.11 82.13 82.14
11721 Scaffolder - Prefabricated, Semi-skilled, Local 75.66 75.92 76.10 76.23 76.33 76.40
11722 Scaffolder - Prefabricated, Semi-skilled, Foreign 62.51 62.04 61.78 61.61 61.42 61.14
11811 Scaffolder - Tubular, Skilled, Local 103.07 102.66 102.48 102.36 102.19 101.89
11812 Scaffolder - Tubular, Skilled, Foreign 81.55 81.23 81.06 80.95 80.82 80.61
11821 Scaffolder - Tubular, Semi-skilled, Local 78.75 78.08 77.74 77.52 77.25 76.83
11822 Scaffolder - Tubular, Semi-skilled, Foreign 60.64 59.17 58.38 57.86 57.26 56.37
11911 Painter - Building, Skilled, Local 101.22 100.58 100.26 100.05 99.79 99.37
11912 Painter - Building, Skilled, Foreign 81.56 81.35 81.24 81.16 81.08 80.96
(continued on next page)

12
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 12 (continued )
Labour Wages Forecasted Rates (RM)

S. No Category of Labour 2020 2021 2022 2023 2024 2025

11921 Painter - Building, Semi-skilled, Local 76.37 75.90 75.66 75.50 75.30 74.99
11922 Painter - Building, Semi-skilled, Foreign 61.31 60.96 60.77 60.65 60.51 60.29
12031 General Construction Worker - Civil, Local 74.84 74.23 73.89 73.67 73.42 73.07
12032 General Construction Worker - Civil, Foreign 54.80 53.70 53.10 52.71 52.26 51.59
Plant and Machine Operators Group
20111 Excavator Operator, Skilled, Local 126.20 124.79 123.89 123.26 122.71 122.15
20122 Excavator Operator, Semi-skilled, Foreign 103.07 101.95 101.20 100.67 100.24 99.84
20211 Pile Rigger, Skilled, Local 120.43 119.12 118.26 117.66 117.16 116.66
20212 Pile Rigger, Skilled, Foreign 93.57 92.10 91.16 90.49 89.92 89.34
20221 Pile Rigger, Semi-skilled, Local 84.13 81.63 80.20 79.23 78.22 76.90
20222 Pile Rigger, Semi-skilled, Foreign 66.17 63.73 62.33 61.39 60.40 59.10
20311 Off Road Truck Operator, Skilled, Local 110.50 108.67 107.56 106.80 106.07 105.21
20312 Off Road Truck Operator, Skilled, Foreign 86.47 85.69 85.20 84.86 84.55 84.22
20321 Off Road Truck Operator, Semi-skilled, Local 75.14 74.18 73.66 73.31 72.92 72.35
20322 Off Road Truck Operator, Semi-skilled, Foreign 60.08 58.63 57.85 57.32 56.73 55.88
20411 Backhoe Loader Operator, Skilled, Local 118.88 117.23 116.15 115.39 114.75 114.12
20412 Backhoe Loader Operator, Skilled, Foreign 92.27 91.00 90.19 89.63 89.14 88.61
20511 Roller Operator, Skilled, Local 105.56 103.94 102.99 102.34 101.70 100.89
20512 Roller Operator, Skilled, Foreign 89.26 88.73 88.37 88.11 87.91 87.74
20521 Roller Operator, Semi-skilled, Local 74.95 73.81 73.17 72.74 72.28 71.63
20522 Roller Operator, Semi-skilled, Foreign 57.55 56.18 55.46 54.98 54.43 53.60
20611 Roller/Compactor Operator, Skilled, Local 104.45 103.61 103.05 102.66 102.33 102.01
20612 Roller/Compactor Operator, Skilled, Foreign 86.58 86.19 85.90 85.68 85.54 85.47
20621 Roller/Compactor Operator, Semi-skilled, Local 72.63 71.26 70.51 70.01 69.45 68.65
20622 Roller/Compactor Operator, Semi-skilled, Foreign 61.87 61.32 61.00 60.79 60.57 60.27
20711 Scrapper Operator, Skilled, Local 110.21 108.92 108.15 107.62 107.11 106.48
20712 Scrapper Operator, Skilled, Foreign 88.21 87.23 86.63 86.22 85.83 85.36
20721 Scrapper Operator, Semi-skilled, Local 75.92 74.62 73.92 73.46 72.93 72.15
20722 Scrapper Operator, Semi-skilled, Foreign 62.12 60.71 59.93 59.41 58.83 58.02
20811 Motor Grader Operator, Skilled, Local 111.51 109.83 108.79 108.08 107.42 106.65
20812 Motor Grader Operator, Skilled, Foreign 91.21 89.61 88.63 87.96 87.33 86.59
20911 Wheel Loader Operator, Skilled, Local 118.07 116.71 115.83 115.21 114.69 114.17
20912 Wheel Loader Operator, Skilled, Foreign 96.16 94.74 93.82 93.18 92.63 92.06
20921 Wheel Loader Operator, Semi-skilled, Local 81.55 78.82 77.27 76.22 75.12 73.64
20922 Wheel Loader Operator, Semi-skilled, Foreign 68.29 66.41 65.33 64.60 63.84 62.84
21011 Paver Operator, Skilled, Local 123.20 121.78 120.82 120.14 119.60 119.13
21012 Paver Operator, Skilled, Foreign 101.70 100.17 99.13 98.39 97.81 97.31
21021 Paver Operator, Semi-skilled, Local 84.22 81.69 80.23 79.24 78.22 76.89
21022 Paver Operator, Semi-skilled, Foreign 71.62 70.02 69.06 68.41 67.77 67.00
21111 Mobile Crane Operator, Skilled, Local 151.92 152.18 152.28 152.33 152.44 152.66
21112 Mobile Crane Operator, Skilled, Foreign 120.37 120.71 120.81 120.86 121.02 121.39
21121 Mobile Crane Operator, Semi-skilled, Local 109.18 108.32 107.79 107.43 107.09 106.70
21122 Mobile Crane Operator, Semi-skilled, Foreign 83.33 82.59 82.16 81.85 81.56 81.21
21211 Crawler Crane Operator, Skilled, Local 140.21 139.61 139.24 138.99 138.76 138.47
21212 Crawler Crane Operator, Skilled, Foreign 109.29 109.23 109.15 109.09 109.08 109.12
21221 Crawler Crane Operator, Semi-skilled, Local 101.58 101.04 100.78 100.62 100.39 100.00
21222 Crawler Crane Operator, Semi-skilled, Foreign 78.70 77.85 77.39 77.08 76.73 76.24
21311 Tower Crane Operator, Skilled, Local 156.58 154.73 153.56 152.73 152.02 151.25
21312 Tower Crane Operator, Skilled, Foreign 121.90 120.36 119.40 118.74 118.14 117.46
21321 Tower Crane Operator, Semi-skilled, Local 113.06 110.45 108.86 107.76 106.73 105.53
21322 Tower Crane Operator, Semi-skilled, Foreign 84.79 82.92 81.83 81.09 80.34 79.37
21411 Forklift Operator, Skilled, Local 100.09 97.49 95.91 94.81 93.79 92.60
21412 Forklift Operator, Skilled, Foreign 83.88 82.41 81.44 80.76 80.19 79.64
21421 Forklift Operator, Semi-skilled, Local 76.49 74.16 72.78 71.84 70.91 69.76
21422 Forklift Operator, Semi-skilled, Foreign 64.14 62.33 61.24 60.49 59.77 58.91
21511 Slinger/Dogger, Skilled, Local 105.34 103.60 102.47 101.67 101.00 100.35
21512 Slinger/Dogger, Skilled, Foreign 89.85 88.18 87.09 86.33 85.69 85.05
21521 Slinger/Dogger, Semi-skilled, Local 80.76 78.82 77.65 76.84 76.08 75.15
21522 Slinger/Dogger, Semi-skilled, Foreign 66.10 64.44 63.44 62.75 62.09 61.31
IBS Installer Group
30111 IBS Precast Concrete Installer, Skilled, Local 105.68 104.94 104.51 104.22 103.93 103.54
30121 IBS Precast Concrete Installer, Semi-skilled, Local 76.97 75.05 74.01 73.32 72.54 71.41
30211 IBS Lightweight Panel Installer, Skilled, Local 107.32 107.88 108.14 108.31 108.54 108.94
30221 IBS Lightweight Panel Installer, Semi-skilled, Local 80.76 80.04 79.63 79.36 79.07 78.66
30311 Lightweight Blockwall Installer, Skilled, Local 103.78 103.10 102.71 102.44 102.17 101.80
30321 Lightweight Blockwall Installer, Semi-skilled, Local 77.01 75.16 74.15 73.48 72.73 71.64
30411 System Formwork Installer, Skilled, Local 108.79 108.83 108.86 108.88 108.89 108.90
30421 System Formwork Installer, Semi-skilled, Local 80.53 78.86 77.93 77.31 76.64 75.70
(continued on next page)

13
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 12 (continued )
Labour Wages Forecasted Rates (RM)

S. No Category of Labour 2020 2021 2022 2023 2024 2025

30511 Roof Truss Installer (Timber), Skilled, Local 104.44 105.00 105.29 105.49 105.72 106.05
30521 Roof Truss Installer (Timber), Semi-skilled, Local 75.84 75.39 75.17 75.03 74.84 74.54
30611 Roof Truss Installer (Light Gauge Steel), Skilled, Local 105.29 105.49 105.59 105.65 105.73 105.88
30621 Roof Truss Installer (Light Gauge Steel), Semi-skilled, Local 81.95 82.36 82.58 82.72 82.89 83.16

1. Selangor region, the hub of construction in Malaysia, was selected as the study area, but the construction rates are most likely to be
the same in all regions.
2. The data from the previous seven years were examined to identify the relation between price increases and inflation.

6. Conclusion

In this study, a construction rate forecasting model was developed to examine the inflation rate relationship with the construction
rates. A model was developed to predict future construction rates based on the influential rate of inflation rate. The number of ob­
servations of building materials prices data were higher and thus were also forecasted using Time Series Analysis, the ARIMA model.
The forecasting was made from 2020 to 2025 with a self-developed and ARIMA models, validated via AIC. The values revealed that the
self-developed model performed better than the ARIMA model. The developed dynamic construction rate forecasting model integrates
past construction rates with the inflation rate and foresees the rate deviation over time. Therefore, its practical implementation should
be made at the initial stage of a project while setting up the project budget which will save the efforts of adjusting the contingency cost.
The model can be utilised in any country’s construction industry with similar project cost overrun issues due to the inflation rate.
Even with several limitations applied, the contribution of this work is effective to the body of knowledge. This study provides
insights into model development by emphasising the connectivity of the inflation rate with the construction industry in affecting its
performance. Besides, a mathematical construction rate forecasting model has been proposed to deal with one of the major concerns of
the construction industry. In a nutshell, this study contributes to theoretical and practical knowledge. As a theoretical contribution, the
studied analysis provides a benchmark for future researchers to pursue the case more deeply and evaluate the other hidden factors that
can impact the project budget. The developed model can further be modified upon assessing the relationships of the construction
industry. While as a practical contribution, this study provides a mathematical construction rate forecasting model which can predict
future prices, that can be embedded into the Bill of Quantities before the tender allotment. The construction industry stakeholders can
foresee the future rates in the present year and make the necessary adjustments to avoid a project being cost overrun.
The following future directions are proposed:

1. The developed dynamic forecasting model integrates past construction rates with the inflation rate and foresees rate deviation over
time. Therefore, practical implementation should be performed at the initial stage of a project while setting up the project budget to
minimise efforts of adjusting the contingency cost.
2. Other forecasting techniques can also be used to draw a comparison to predict the construction rates based on the influence of the
inflation rate.
3. Other influential factors, such as oil prices, should be examined to observe the influence of construction rates in deviation.

Funding

Not applicable.

Data availability

1. Sharing research data helps other researchers evaluate your findings, build on your work and to increase trust in your article. We
encourage all our authors to make as much of their data publicly available as reasonably possible. Please note that your response to the
following questions regarding the public data availability and the reasons for potentially not making data available will be available
alongside your article upon publication.
Has data associated with your study been deposited into a publicly available repository?
Ans: No.

CRediT authorship contribution statement

Muhammad Ali Musarat: Writing – original draft, Validation, Software, Methodology, Investigation, Formal analysis, Data
curation, Conceptualization. Wesam Salah Alaloul: Writing – review & editing, Visualization, Supervision, Resources, Project
administration, Methodology, Data curation, Conceptualization. M.S. Liew: Writing – review & editing, Visualization, Supervision,
Resources, Project administration, Funding acquisition.

14
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 13
Labour wages validation.
S. No Forecasted Actual S. No Forecasted Actual S. No Forecasted Actual

Construction Workers Group

10131 64.50 68.20 10132 53.05 55.00 10211 90.59 94.50
5.73% 3.68% 4.32%
10212 72.19 85.50 10221 68.08 79.00 10222 58.48 67.60
18.44% 16.04% 15.60%
10311 98.95 110.50 10312 76.58 79.00 10321 71.05 76.00
11.67% 3.16% 6.97%
10322 58.66 64.60 10411 111.13 111.50 10412 84.38 85.60
10.13% 0.34% 1.45%
10421 76.80 80.50 10422 64.24 67.80 10511 121.98 135.00
4.82% 5.55% 10.67%
10512 94.65 90.00 10521 90.13 94.00 10522 70.96 66.10
− 4.91% 4.29% − 6.85%
10611 108.06 106.00 10612 83.11 74.00 10621 80.63 91.00
− 1.90% − 10.96% 12.87%
10622 63.39 61.60 10711 104.51 113.00 10712 79.59 94.00
− 2.82% 8.12% 18.11%
10721 77.53 90.00 10722 63.58 63.60 10811 126.45 127.00
16.09% 0.03% 0.43%
10812 94.02 92.50 10821 83.86 102.00 10822 65.49 73.00
− 1.61% 21.64% 11.46%
10911 108.12 110.00 10912 84.85 84.00 10921 79.76 84.00
1.74% − 1.01% 5.32%
10922 60.88 68.00 11011 118.59 123.00 11012 92.83 85.50
11.70% 3.72% − 7.90%
11021 87.00 77.50 11022 69.37 66.50 11111 125.28 117.60
− 10.92% − 4.13% − 6.13%
11112 97.52 83.00 11121 87.64 81.00 11122 72.04 64.20
− 14.88% − 7.57% − 10.88%
11211 114.01 111.50 11212 90.51 84.00 11221 82.61 79.00
− 2.20% − 7.20% − 4.37%
11222 64.91 65.00 11311 113.78 113.50 11312 89.04 84.00
0.14% − 0.24% − 5.66%
11321 82.88 77.00 11322 69.86 64.60 11421 98.62 99.50
− 7.10% − 7.54% 0.90%
11422 79.40 75.60 11511 135.46 138.64 11512 105.80 107.55
− 4.78% 2.34% 1.65%
11611 159.48 173.65 11612 121.84 112.91 11711 99.75 101.80
8.88% − 7.33% 2.05%
11712 82.03 90.50 11721 75.66 75.50 11722 62.51 72.00
10.32% − 0.21% 15.18%
11811 103.07 101.50 11812 81.55 78.00 11821 78.75 80.50
− 1.52% − 4.36% 2.23%
11822 60.64 64.60 11911 101.22 110.00 11912 81.56 83.00
6.53% 8.67% 1.76%
11921 76.37 80.00 11922 61.31 64.50 12031 74.84 88.00
4.75% 5.21% 17.58%
12032 54.80 62.62 20111 126.20 134.45 20122 103.07 97.10
14.26% 6.54% − 5.79%
20211 120.43 119.50 20212 93.57 95.00 20221 84.13 103.50
− 0.77% 1.52% 23.02%
20222 66.17 82.50 20311 110.50 129.50 20312 86.47 87.50
24.67% 17.19% 1.19%
20321 75.14 87.50 20322 60.08 72.50 20411 118.88 134.65
16.44% 20.67% 13.26%
20412 92.27 104.40 20511 105.56 118.70 20512 89.26 101.30
13.14% 12.45% 13.49%
20521 74.95 92.40 20522 57.55 65.50 20611 104.45 116.00
23.28% 13.82% 11.05%
20612 86.58 90.00 20621 72.63 91.50 20622 61.87 65.50
3.95% 25.98% 5.86%
20711 110.21 128.20 20712 88.21 103.00 20721 75.92 92.00
16.32% 16.76% 21.18%
20722 62.12 72.00 20811 111.51 124.20 20812 91.21 94.50
15.90% 11.38% 3.61%
20911 118.07 124.90 20912 96.16 104.30 20921 81.55 95.00
5.79% 8.46% 16.50%
20922 68.29 69.00 21011 123.20 138.00 21012 101.70 108.30
(continued on next page)

15
M.A. Musarat et al. Heliyon 10 (2024) e26037

Table 13 (continued )
S. No Forecasted Actual S. No Forecasted Actual S. No Forecasted Actual

1.04% 12.02% 6.49%

21021 84.22 103.90 21022 71.62 80.00 21111 151.92 151.50
23.37% 11.70% − 0.28%
21112 120.37 110.80 21121 109.18 123.50 21122 83.33 76.00
− 7.95% 13.12% − 8.79%
21211 140.21 153.30 21212 109.29 108.00 21221 101.58 112.50
9.34% − 1.18% 10.75%
21222 78.70 73.00 21311 156.58 188.70 21312 121.90 134.00
− 7.24% 20.51% 9.92%
21321 113.06 135.50 21322 84.79 101.50 21411 100.09 106.10
19.85% 19.71% 6.00%
21412 83.88 77.00 21421 76.49 84.50 21422 64.14 67.50
− 8.21% 10.47% 5.24%
21511 105.34 106.34 21512 89.85 88.00 21521 80.76 88.19
0.95% − 2.06% 9.20%
21522 66.10 69.00 30111 105.68 110.50 30121 76.97 81.50
4.38% 4.56% 5.88%
30211 107.32 97.50 30221 80.76 83.00 30311 103.78 103.50
− 9.15% 2.77% − 0.27%
30321 77.01 83.00 30411 108.79 103.50 30421 80.53 83.00
7.78% − 4.87% 3.07%
30511 104.44 132.00 30521 75.84 91.00 30611 105.29 114.00
26.39% 19.99% 8.27%
30621 81.95 84.00
2.51%
Overall Deviation ¼ 5.69%

Table 14
Forecasted machinery hire rates.
Machinery Hire Rates Forecasted Rates (RM)

S. No Category 2020 2021 2022 2023 2024 2025

1.1 Hydraulic Excavator. Komatsu, PC200-7 9483.03 9421.04 9377.15 9345.59 9322.46 9305.17
1.2 Hydraulic Excavator, Komatsu, PC300-7 16555.17 16511.96 16481.43 16459.49 16443.36 16431.2
1.3 Hydraulic Excavator, Komatsu, PC400LC-7 24505.25 23985.24 23615.78 23349.83 23156.12 23013.53
1.4 Hydraulic Excavator, Hitachi, ZAXIS 120 8538.666 8429.148 8351.581 8295.79 8254.943 8224.455
1.5 Tracker Excavator, Sumitomo, SH120-3 8518.022 8407.708 8329.552 8273.333 8232.194 8201.531
2.1 Bulldozer, Komatsu, D65E-12 14429.55 14266.63 14150.87 14067.54 14006.85 13962.18
3.1 Motor Grader, Caterpillar, 140H Standard 14766.79 14394.45 14129.95 13939.56 13800.85 13698.66
4.1 Lorry, Hino, BDM 10000 kg 15528.77 15335.96 15199.14 15100.68 15028.82 14975.62
4.2 Lorry, Hino, BDM 20000 kg 21409.77 21114.5 20904.79 20753.85 20643.85 20562.73
4.3 Lorry, Nissan, BDM 3000 kg 12597.74 12372.62 12212.81 12097.8 12013.91 11951.91
4.4 Lorry, Nissan, BDM 5000 kg 14425.89 14093.58 13857.55 13687.66 13563.85 13472.59
4.5 Lorry, Nissan, BDM 10000 kg 16455.81 16225.91 16062.75 15945.33 15859.66 15796.28
5.1 Mobile Crane, Kato, NK200H II 12066.1 11851.29 11698.57 11588.62 11508.62 11449.9
5.2 Mobile Crane, Kato, NK450B 23287.36 22846.82 22533.83 22308.53 22144.43 22023.59

Table 15
Demographic profile of respondents.
Respondent Qualification Year of Experience Organisation Type Business Nature

1 B.Sc Civil Engineering 10 Private Contractor

2 B.Sc Civil Engineering 10 Public Client
3 B.Sc Electrical Engineering 8 Private Consultant
4 B.Sc Civil Engineering 16 Private Client
5 PhD in Civil Engineering 29 Public Client
6 PhD in Civil Engineering 18 Public Academic
7 B.Sc Civil Engineering 26 Private Contractor
8 PhD in Civil Engineering 22 Public Consultant
9 B.Sc Civil Engineering 40+ Private Client
10 B.Sc Civil Engineering 31 Private Client
11 B.Sc Civil Engineering 3 Private Client
12 B.Sc Civil Engineering 3 Private Client
13 B.Sc Civil Engineering 14 Private Client
14 B.Sc Civil Engineering 20 Private Client

16
M.A. Musarat et al. Heliyon 10 (2024) e26037

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Acknowledgement

The authors would like to thank the Universiti Teknologi PETRONAS (UTP) for the support from the (cost centre 015PBC-004)
awarded to Wesam Salah Alaloul.

List of Abbreviations

Abbreviation Definition
ES Exponential Smoothing
AR Autoregressive
MA Moving Average
ARMA Autoregressive Moving Average
I Integrated
ARIMA Autoregressive Integrated Moving Average
ANN Artificial Neural Network
MSE Mean Square Error
MAPE Mean Absolute Percentage Error
r Correlation Coefficient
SAR Seasonal Autoregressive
SMA Seasonal Moving average
AIC Akaike’s Information Criterion

References

[1] A. Aljohani, D. Ahiaga-Dagbui, D. Moore, Construction projects cost overrun: what does the literature tell us? Int. J. Innov., Manag. Technol. 8 (2) (2017) 137.
[2] M. El-Ahwal, S.S. El-Attar, W.A. Abdel-Hafez, Factors leading to cost overrun occurrence in construction projects, Port-Said Eng. Res. J. 20 (1) (2016) 71–77.
[3] M. Enrica, H. Purba, A. Purba, Risks leading to cost overrun in construction projects: a systematic literature review, Adv. Res. Civil Eng. 3 (1) (2021) 43–60.
[4] L.M. Khodeir, A. El Ghandour, Examining the role of value management in controlling cost overrun [application on residential construction projects in Egypt],
Ain Shams Eng. J. 10 (3) (2019) 471–479.
[5] X. Mashwama, C. Aigbavboa, D. Thwala, Investigation of construction stakeholders’ perception on the effects & cost of construction dispute in Swaziland,
Procedia Eng. 164 (2016) 196–205.
[6] A. Firouzi, W. Yang, C.-Q. Li, Prediction of total cost of construction project with dependent cost items, J. Construct. Eng. Manag. 142 (12) (2016) 04016072.
[7] F.P. Catalão, C.O. Cruz, J.M. Sarmento, The determinants of cost deviations and overruns in transport projects, an endogenous models approach, Transport Pol.
74 (2019) 224–238.
[8] A.H. Memon, I.A. Rahman, A.A. Azis, Assessing causal relationships between construction resources and cost overrun using PLS path modelling focusing in
southern and central region of Malaysia, J. Eng. Technol. 4 (1) (2013) 67–78.
[9] A.H. Memon, I.A. Rahman, Budget overrun issues in construction projects of southern part of Malaysia, Int J Civ Eng Built Environ 1 (2014) 1–6.
[10] S. Durdyev, S. Ismail, N.A. Bakar, Factors causing cost overruns in construction of residential projects: case study of Turkey, Int. J. Sci. Manag. 1 (1) (2012)
3–12.
[11] S. Al-Azzawi, A.M. Hasan, A new 4D hidden hyperchaotic system with higher largest lyapunov exponent and its synchronization, Int. J. Math., Stat. Comput. Sci.
2 (2024) 63–74.
[12] S. Siami-Namini, N. Tavakoli, A.S. Namin, A comparison of ARIMA and LSTM in forecasting time series, in: 2018 17th IEEE International Conference on
Machine Learning and Applications (ICMLA), IEEE, 2018.
[13] M.B.A. Rabbani, M.A. Musarat, W.S. Alaloul, M.S. Rabbani, A. Maqsoom, S. Ayub, H. Bukhari, M. Altaf, A comparison between seasonal autoregressive
integrated moving average (SARIMA) and exponential smoothing (ES) based on time series model for forecasting road accidents, Arabian J. Sci. Eng. 46 (11)
(2021) 11113–11138.
[14] G.E. Box, G.M. Jenkins, G.C. Reinsel, G.M. Ljung, Time Series Analysis: Forecasting and Control, John Wiley & Sons, 2015.
[15] W.S. Alaloul, M.A. Musarat, M.B.A. Rabbani, Q. Iqbal, A. Maqsoom, W. Farooq, Construction sector contribution to economic stability: Malaysian GDP
distribution, Sustainability 13 (9) (2021) 5012.
[16] C.-C. Wang, A comparison study between fuzzy time series model and ARIMA model for forecasting Taiwan export, Expert Syst. Appl. 38 (8) (2011) 9296–9304.
[17] M. Al Wadia, M.T. Ismail, Selecting wavelet transforms model in forecasting financial time series data based on ARIMA model, Appl. Math. Sci. 5 (7) (2011)
315–326.
[18] G.P. Zhang, Time series forecasting using a hybrid ARIMA and neural network model, Neurocomputing 50 (2003) 159–175.
[19] F.-M. Tseng, H.-C. Yu, G.-H. Tzeng, Combining neural network model with seasonal time series ARIMA model, Technol. Forecast. Soc. Change 69 (1) (2002)
71–87.
[20] L. Wang, H. Zou, J. Su, L. Li, S. Chaudhry, An ARIMA-ANN hybrid model for time series forecasting, Syst. Res. Behav. Sci. 30 (3) (2013) 244–259.
[21] E.S. Gardner Jr., Exponential smoothing: the state of the art—Part II, Int. J. Forecast. 22 (4) (2006) 637–666.
[22] E.S. Gardner Jr., Exponential smoothing: the state of the art, J. Forecast. 4 (1) (1985) 1–28.
[23] D.C. Montgomery, L.A. Johnson, J.S. Gardiner, Forecasting and Time Series Analysis, McGraw-Hill Companies, 1990.
[24] E. Ostertagová, O. Ostertag, The simple exponential smoothing model, in: The 4th International Conference on Modelling of Mechanical and Mechatronic
Systems, Technical University of Košice, Slovak Republic, 2011. Proceedings of conference.
[25] A. Aczel, Complete Business Statistics, Irwin, 1989. ISBN 0-256-05710-8.

17
M.A. Musarat et al. Heliyon 10 (2024) e26037

[26] R.J. Hyndman, A.B. Koehler, R.D. Snyder, S. Grose, A state space framework for automatic forecasting using exponential smoothing methods, Int. J. Forecast. 18
(3) (2002) 439–454.
[27] K. Ord, Charles Holt’s report on exponentially weighted moving averages: an introduction and appreciation, Int. J. Forecast. 20 (1) (2004) 1–3.
[28] R.G. Brown, Statistical Forecasting for Inventory Control, McGraw/Hill, 1959.
[29] P.R. Winters, Forecasting sales by exponentially weighted moving averages, Manag. Sci. 6 (3) (1960) 324–342.
[30] B. Billah, M.L. King, R.D. Snyder, A.B. Koehler, Exponential smoothing model selection for forecasting, Int. J. Forecast. 22 (2) (2006) 239–247.
[31] J.K. Ord, A.B. Koehler, R.D. Snyder, Estimation and prediction for a class of dynamic nonlinear statistical models, J. Am. Stat. Assoc. 92 (440) (1997)
1621–1629.
[32] M.A. Musarat, W.S. Alaloul, M. Liew, Impact of inflation rate on construction projects budget: a review, Ain Shams Eng, J 12 (1) (2021) 407–414.
[33] Writer, H.A., IMPACT OF INFLATION ON GOVERNMENT SPENDING IN NIGERIAN ECONOMY (1981-2013)..
[34] W.S. Alaloul, M.A. Musarat, M.B.A. Rabbani, M. Altaf, K.M. Alzubi, M. Al Salaheen, Assessment of economic sustainability in the construction sector: evidence
from three developed countries (the USA, China, and the UK), Sustainability 14 (10) (2022) 6326.
[35] M. Feldkircher, P.L. Siklos, Global inflation dynamics and inflation expectations, Int. Rev. Econ. Finance. 64 (2019) 217–241.
[36] K. Oikawa, K. Ueda, The optimal inflation rate under Schumpeterian growth, J. Monetary Econ. 100 (2018) 114–125.
[37] E. Gulsen, H. Kara, Measuring inflation uncertainty in Turkey, Central Bank Rev. 19 (2) (2019) 33–43.
[38] E. Osabuohien, E. Obiekwe, E.S. Urhie, R. Osabohien, Inflation rate, exchange rate volatility and exchange rate pass-through Nexus: the Nigerian experience,
J. Appl. Econ. Sci. 2 (56) (2018) 574–585.
[39] K. Szafranek, Bagged neural networks for forecasting Polish (low) inflation, Int. J. Forecast. 35 (3) (2019) 1042–1059.
[40] A.N. Haslinda, T.W. Xian, K. Norfarahayu, R.M. Hanafi, H.M. Fikri, Investigation on the factors influencing construction time and cost overrun for high-rise
building projects in penang, in: Journal of Physics: Conference Series, IOP Publishing, 2018.
[41] M.A. Musarat, W.S. Alaloul, M. Liew, A. Maqsoom, A.H. Qureshi, The effect of inflation rate on CO2 emission: a framework for Malaysian construction industry,
Sustainability 13 (3) (2021) 1562.
[42] C.F. Tang, The effect of real wages and inflation on labour productivity in Malaysia, Int. Rev. Appl. Econ. 28 (3) (2014) 311–322.
[43] W.S. Alaloul, M.A. Musarat, M. Liew, A.H. Qureshi, A. Maqsoom, Investigating the impact of inflation on labour wages in Construction Industry of Malaysia, Ain
Shams Eng, J 12 (2) (2021) 1575–1582.
[44] W.S. Alaloul, M.A. Musarat, M. Liew, A.H. Qureshi, Influence of inflation rate on machinery hire rates in construction industry, J. Civil Eng., Sci. Technol. 12 (1)
(2021) 39–45.
[45] M.A. Musarat, W.S. Alaloul, A.H. Qureshi, M. Altaf, Inflation rate and construction materials prices: relationship investigation, in: 2020 International Conference
on Decision Aid Sciences and Application (DASA), IEEE, 2020.
[46] Y.A. Olawale, M. Sun, Cost and time control of construction projects: inhibiting factors and mitigating measures in practice, Construct. Manag. Econ. 28 (5)
(2010) 509–526.
[47] D. Leavitt, S. Ennis, P. McGovern, The Cost Escalation of Rail Projects: Using Previous Experience to Re-evaluate the Calspeed Estimates, 1993.
[48] R.F. Aziz, Factors causing cost variation for constructing wastewater projects in Egypt, Alex. Eng. J. 52 (1) (2013) 51–66.
[49] S. Goyal, Effect of inflation on project costs in Rajasthan, in: Proceedings of SARC-CAN International Conference, University Rajasthan, New Delhi, India, 2017.
12th March.
[50] G.A. Niazi, N. Painting, Significant factors causing cost overruns in the construction industry in Afghanistan, Procedia Eng. 182 (2017) 510–517.
[51] H. Alinaitwe, R. Apolot, D. Tindiwensi, Investigation into the causes of delays and cost overruns in Uganda’s public sector construction projects, J. Constr. Dev.
Ctries. (JCDC) 18 (2) (2013) 33.
[52] O. Oghenekevwe, O. Olusola, U.S. Chukwudi, An assessment of the impact of inflation on construction material prices in Nigeria, PM World J. 3 (4) (2014) 1–22.
[53] C. Kaliba, M. Muya, K. Mumba, Cost escalation and schedule delays in road construction projects in Zambia, Int. J. Proj. Manag. 27 (5) (2009) 522–531.
[54] L. Le-Hoai, Y. Dai Lee, J.Y. Lee, Delay and cost overruns in Vietnam large construction projects: a comparison with other selected countries, KSCE J. Civ. Eng. 12
(6) (2008) 367–377.
[55] A. Enshassi, J. Al-Najjar, M. Kumaraswamy, Delays and cost overruns in the construction projects in the Gaza Strip, J. Financ. Manag. Prop. Construc. 14 (2)
(2009) 126–151.
[56] P. Bakhshi, A. Touran, An overview of budget contingency calculation methods in construction industry, Procedia Eng. 85 (2014) 52–60.
[57] A.U. Rahman, M. Saeed, M.A. Mohammed, A. Majumdar, O. Thinnukool, Supplier selection through multicriteria decision-making algorithmic approach based
on rough approximation of fuzzy hypersoft sets for construction project, Buildings 12 (7) (2022) 940.

18