Smart fleet management system based on multi-

agent systems : Mining context

Hajar BNOUACHIR1, Hicham MEDROMI6 Mourad ZEGRARI3

Structural Engineering, Intelligent Systems and Electrical
Engineering research laboratory (LRI), System Architecture
Energy Laboratory, Ecole Nationale Supérieure des Arts et
Team (EAS), National and high school of electricity and
Métiers, Hassan II University of Casablanca, Innovation
mechanic (ENSEM), Hassan II University of Casablanca.
Lab for Operations, Mohammed VI Polytechnic University
Research Foundation for Development and Innovation in
Benguerir, Morocco, Casablanca Morocco
Science and Engineering.
Casablanca, Morocco.

Meryiem CHERGUI2 , Aziza CHAKIR4

Engineering Research Laboratory (LRI), System
Laurent DESHAYES5
Innovation Lab for Operations,
Architecture (EAS), National and High School of Electricity
Mohammed VI Polytechnic University
and Team Mechanic (ENSEM), Hassan II University of
Benguerir, Morocco
Casablanca, Casablanca, Morocco

Abstract—Fleet management has long been limited to vehicle actions carried out and the budgets committed must be
tracking. With Big Data, this sector is changing dramatically and monitored, both for quantitative and qualitative objectives.
opening up new horizons for companies. It aims to optimize the
use of vehicles, reduce costs and environmental impact, and Generally, Fleet management system (FMS) is a term
improve driver safety and service to the customer. However, used to designate a wide range of solutions for the various fleet
their traditional control architecture is becoming a critical management problems related to the fleet in the fields of
bottleneck in open and dynamic environments, where scalability
transport, distribution and logistics, it includes targeted
and autonomy are key factors in their success.
In this article, we propose an intelligent distributed fleet planning, as well as supervision and control of fleet operations
management system architecture for an open pit mine that allows according to the available transport resources and application
optimizing the production and efficiency of the mine based on constraints. The objective of the FMS is to reduce risks,
real-time data processing, according to the users' requirements,
in order to obtain a system that optimally manages and plans the increase the quality of service and improve the operational
fleet of mining machines within a digital mine; this instantaneous efficiency of a fleet while minimizing its costs.
management allows visualizing and knowing the real situation of The Fleet Management System (FMS) is a modern
the mine and following its short and long term objectives. This technological solution for managing the fleet vehicles in terms
will provide a relevant decision support system.
of automating and optimizing business processes and linking
Keywords—Fleet management system, open pit mine, all business areas such as dispatching, purchasing,
Monitoring, architectures, multi-agent system, real time communications, navigation, accounting and finance in a single
system system.
Mining projects, especially open-pit mines, are capital-
I. INTRODUCTION intensive enterprises with high operating costs. Approximately
The monitoring of the company's activity in real time has 50% of the operating costs in open-pit mines [1]and even 60%
many advantages and benefits that should not be overlooked. in large open-pit mines are spent on transport and material
This instantaneous management makes it possible to visualize handling[1] .Of all material handling operations in open-pit
and know the real situation of the company as well as to follow mines, this is the most expensive to operate[2].
its objectives, key figures and performance indicators. It Therefore, the integration of the fleet management system in
facilitates decision-making, manages its daily activities more mines, and particularly in open pit mines, aims at optimizing
efficiently and improves its reactivity and competitiveness. production and mine efficiency based on real-time data.
Regardless of the structure size, any good manager must be Specifically, the FMS aims to maximize mine production,
equipped with the appropriate resources to effectively manage minimize inventory handling, feed the mill at the expected rate,
his or her activity. This means that the positioning of the and meet blending constraints[3] .
 In this paper, we will first present the state of the art of quality of administration and increase the operational
fleet management systems as well as the use of artificial efficiency of the fleet while keeping costs down.
intelligence in these systems; we will also introduce a model The implementation of operational research techniques as
for evaluating fleet management systems based on ISA 95. We the main means of addressing the high cost of transporting
will then address the problems of FMS in open pit mines equipment to open-pit mines, has attracted researchers and laid
before proposing a general intelligent FMS architecture. We the foundation for fleet management systems (FMS) used in
will end this article with a conclusion and perspectives. open-pit mines.
They considered the FMS as a real time decision-making
II. FLEET MANAGEMENT SYSTEMS AND ARTIFICIAL
system for materials handling in an open pit mine. FMS obtains
INTELLIGENCE
the required information about mining operations from the
A. Fleet management system: database and makes decisions accordingly. These decisions are
In general, recent improvements in communication and then implemented in the operation and the FMS is recalled
information technology provide new streams of real-time data whenever a new decision is required.
that can be used to define new practical fields of application in Figure 1 shows the communication between open-pit mining
transport across all industrial sectors. and the FMS

Indeed, at any given moment, a carrier[4] can for example:

1. Know the position of each of its vehicles.
2. Obtain information about traffic congestion.
3. To be in contact with each of the vehicles in its fleet in a
bidirectional manner (transmission of information to drivers,
confirmation of execution of a mission from the driver, etc.).
4. communicate to customers the status of their requests.

Real-time fleet management problems are classified as

dynamic transport problems. As a reminder of the definition of
a dynamic problem, we mention the one formulated by Powell.
Fig. 1. Schematic of the surface mining operation and how it communicates
Jaillet and Odoni[5]: with the FMS[6]
"A problem is dynamic if decisions have to be made before all One of the main decisions to be taken by the FMS is to
information is known and changed when new information is send trucks by shovel. This decision must be made in a way
denied. that meets production requirements with the minimum variance
and maximizes the productivity of the active equipment,
The problem of real-time fleet management therefore consists including shippers and carriers. Although, since the 1960s,
of: different strategies and heuristic methods have been used to
• Allocate a new requirement or any demand for service make the crucial decision on truck dispatching, none of them is
made to the transporter that requires the intervention of at able to achieve all of the above objectives simultaneously[6].
least one vehicle of the fleet to carry it out to a given cycle
while previously established cycles are running. We In general, models for solving fleet management
designate by Cycle a sequence of missions carried out or problems can be divided into two families: fixed and dynamic
planned for a vehicle over a given time horizon. models. We will focus on fleet management in open pit mines.
• Ensure that the cycles run correctly with regard to the Consequently, the allocation of trucks to the open pit mine
planned actions that one or more vehicles must carry out in includes two types of assignment: Fixed Assignment and
order to achieve one or more objectives {move, load. Dynamic Assignment:
unload, reposition, etc.).
Overall, Fleet Management System (FMS) is a term used • Fixed distribution: In this method, a group of trucks is
to refer to a wide range of responses for various applications in assigned to a fixed transport route (special shovel) and this
transport, broadcasting and coordination. It includes the condition will remain unchanged until the end of the shift.
organization, management and control of fleet activities, as The routes to which the trucks have been assigned will not
indicated by accessible vehicle resources and application change until a shovel breaks down or a critical event
requirements. The FMS aims to reduce hazards, improve the occurs. In other words, it can be said that a fixed
 dispatching method is a static method. The trucks assigned • Allocation planning[13]: minimizes the total number of
to the routes must work on the same route during the shift trucks required using the sim-plex method, the model is in
according to several criteria, such as production needs, real time, the flow rate of each road is based on the volume
availability of trucks in the fleet, etc. The trucks assigned of material rather than the number of trucks. However, it is
to the routes must work on the same route during the shift only appropriate when some variables are involved and it
according to several criteria, such as production needs, is non-negative constraints for all variables.
availability of trucks in the fleet, etc. The trucks assigned • Real-time Dispatching[14]: Minimizes button waste by
to the routes must work on the same route during the shift applying the dynamic programming method, it propose a
according to several criteria, such as production needs, progressive time horizon when the assignment order is
availability of trucks in the fleet, etc. Classical methods in required, Undercarriage and over trucking conditions are
which the principles of queue theory are used as well as taken into account, but the definition of a progressive time
programming are among the different methods applied for horizon for an assignment order is indispensable, taking
the fixed assignment of trucks. All of these methods aim to into account sub-campionage and over-trucking
determine the optimal number of trucks allocated to each conditions.
excavator[7]. There are many limits of the current algorithms, namely:
• Dynamic allocation: The results of research conducted by
researchers in this field have indicated that a fixed • The link between the strategic plan and operations:
allocation method is not useful and efficient for scheduling the proposed fleet management systems do not allow
transportation in large mines[7]. In dynamic truck both short-term and long-term objectives to be
allocation, a number of trucks available in the fleet are achieved,
assigned to a specific shovel at the beginning of the shift. • Most models are deterministic: they assume a constant
However, these trucks, instead of being assigned to a average grade for each size front and do not cover the
single shovel or route during the shift, will be reassigned entire life of the mine,
from the dispatch system each time after loading and • The current systems do not take into account the loss
unloading at the hoppers[3]. of tones associated with the movement of excavators
Several studies have shown that by adopting the flexible from one level to another, they present one of the
allocation method, loading and transport capacity can be main factors of deviation from the production rate,
significantly improved by a high percentage. Olson et al[8]
showed that using dynamic allocation of haulage trucks in B. Artificial intelligence for fleet management systems:
the Bougainville copper mine, the productivity of the mine Artificial intelligence (AI) is omnipresent nowadays in
operation was improved by 13%. They also demonstrated many technological applications (apps, websites, gaming
predictions, services…etc) It’s a way to give accurate user
a 10-15% increase in gold productivity at the Barrick recommendations and change its experience in many fields
Goldstrike Goldstrike mine, a 10% improvement in LTV [15].
iron mine productivity and a 10% increase in coal Fleet management has revolutionized AI. The priority
production at the Quintette mine[8]. A comparative study given to commercial results and driver safety preserving cost
conducted by Kolonja and Mutmansky [9] on the and efficiency introduce intelligent fleet management systems.
difference between a fixed and a variable allocation The use of smartphones and devices recommending the
best way to take is for the average driver synonym of AI
method demonstrated that the use of variable allocation
presence. Drivers’ lives are easier thanks to complex traffic
leads to a significant increase in the productivity of the apps that combine GPS and artificial intelligence [16]. There
mine operation. are many benefits of AI-based applications for fleet
management such as route recommendations, road risk data
The modular mining system revealed models[10] [11]and analysis and driver coaching. In fact, earlier technology failed
algorithms, namely: to provide the accuracy, efficiency, convenience as AI apps
do. As a result, it is becoming the optimal and safer solution to
• Shortest path[12] [3]: it aims at optimizing travel time transport goods and services in general and in mining field as
well.
using the Dijkstra method, its advantages is that the
We can define AI fleet management as the use of artificial
algorithm does not need to study all contours and it has an intelligence to manage fleet operations, it rationalizes fleet
order of n2 so that it is sufficiently efficient to use for manager tasks by eliminating human error in transportation
relatively large applications. But the model takes a long processes[17] .
time, it has failures in case of negative edges.
 With AI recommendations in FMS, drivers, managers, On the other hand, we carried out within the framework
and mechanics can make correct and optimal decisions to of this project, a needs study within the OCP mining company
improve the fleet long-term performance. It also allows drivers (Benguerir mine), and we identified the following
autonomy during each transport cycle. The table below specifications:
presents some key aspects of fleet management that AI can • Integrated management of the production chain;
optimize: • Optimization of equipment performance and
elimination of losses;
TABLE I. IA FOR FMS OPTIMIZATION
• Improved maintenance efficiency;
IA ASPECT USES RESULTS • Ensure equipment security.
Real-Time Fleet Collect data for predictive • Prioritization
Analytics analytics: opportunities
• traffic and risks According to the previous specifications and the
• road conditions, • Make better evaluation of the FMS conducted in section 2.3 and given the
• environmental hazards, routes benefits of AI in this context, we propose a generic
• real-time weather, • Schedules, architecture of an Intelligent Fleet Management System for an
• mechanical faults maintenance
open pit mine.
delivery
• Dispatch fleet
outcomes and
IV. THE ARCHITECTURE PROPOSAL
activities The real-time aspect in the FMS is necessary to deal with
arrangements
unforeseen events. It makes it possible to detect deviations
• Stay prepared
for any from the initial allocation plan and adjust the schedule by
unexpected offering immediate and efficient re-routing. [19].
events.
In this section, we present our fleet management proposal with
a description of an FMS intelligent architecture (see Figure 4).
Better Repair & Predictive maintenance • Accurate self-
Maintenance Fleets state at all times diagnostics The intelligent FMS architecture includes three basic layers:
Decisions • Saving a routine
maintenance
cost IoT layer contains intelligent sensors for both autonomous and
Fleet Integration • Planning, maintenance and • Integrating manual vehicles in the open pit mine. Every sensor is
monitoring operations every connected to an intelligent agent as software and these agents
department on a are linked to control device agent to manage them and to
single platform
announce the real time status of the mine fleet for different
• Save time and
costs stakeholders.
• Cohesive fleet
Simpler • Knowledge Management • Simplify the on
Recruitment • The most qualified drivers’ boarding
Process recommendation process

III. PROBLEMATIC
The proposed models must be as close as possible to
reality, taking into account the heterogeneity of trucks and the
complete modelling of an open-pit mining operation. FMS
should also determine the shortest dynamic routes taking into
account the current location of the truck, its next destination
and the time required to reach the objective in any current
traffic congestion.
Otherwise, to identify FMS users’ needs and to define its
main attributes in real time, a process was initiated by
Fig. 2. SMART fleet management system architecture
Zeimpekis and Vasileios[18] .The results revealed that the
mining fleet management systems must meet logistics and In fact, control device agent is responsible for detecting new
distribution requirements such as: device (new vehicle IoT sensor), update its status in FMS
• Real-time tracking of vehicle location mapping, suspend a device and detect maintenance needs,
• Generation of vehicle and distribution system dangers and special behaviors.).
performance reports. Indeed, on each vehicle, IoT sensor is implemented and
• Generation of proof of delivery at the end of loading connected to open pit mine network. These sensors allow the
and shipment. acquisition and transmission of vehicles’’ information’s such
as the position of the machines, their speed and their V. IMPLEMENTATION
conditions. They are also aware of each other’s existence. One As for expected tools for the proposed intelligent fleet
more important thing some vehicles in the mine need more management system, we use mainly Internet of Things (IoT)
intelligent sensors for specific mission or critical situations technology. IoT refers to a network of actuators and sensors
and each vehicle in the fleet is aware and reacts to the entire continuously collecting data from
fleet’s needs. A disruption engine determines if a fleet vehicle IoT is a network of sensors continuously collecting
will be critically late for a scheduled trip and allocate environmental data. In the open pit fleet management system,
alternative vehicles, as required. the IoT collects data for analysis. It ensures the transparent
sharing of information between all stakeholders in the supply
Processing layer represents the fleet coordination entities. In chain, such as drivers, production managers and dispatchers.
fact, in this level, we have: For our intelligent fleet management system, we use the three
• Fleet control entity that monitors the operating states main technologies of IoT:
and positions of the vehicles through the communication • Wireless communication (4G, Bluetooth, WiFi) to
with the device control agent. It informs stakeholders transmit the relevant information between the
and the other agents as well of any changes in the fleet different vehicles of the mine.
that would require an adjustment of the assignment of
• Global positioning system (GPS) for precise location
tasks. It has fleet reference states to be able to draw up a
tracking in real time.
real-time, vigilant and continuous comparison
• Integrated diagnostic analyzers (such as OBDII and
• Dispatcher Agent: once executed, this entity recalculates
J1939) for self-diagnosis and reporting.
the overall allocation of tasks for mining vehicles. It has
two dispatching algorithms based on a set of assignment
criteria: Short-term allocation algorithm and a long-term the smart FMS also has cameras to ensure that video data
allocation algorithm. The dispatcher is one of the most can be captured, analyzed and accessible at all times, allowing
intelligent entities in our FMS. It is up to it to define the better study of driver behaviour, road conditions or hazards.
mining missions by taking into consideration the All the more, machine learning technology allows fleets to
strategic decisions and the real state of the fleet. learn data collected over time and make managed adjustments
Dispatcher agent also executes optimized dispatching in based on that data in the performance management entity.
collaboration with performance and maintenance
systems to research fleet vehicle that should go next and The proposed intelligent FMS with all the components and
in which itinerary. technologies used is able to perform the following tasks:
• Maintenance system manages vehicles under • Collection of precise mining transport data and its
maintenance. It allows updating their status and gives transmission to other devices.
information about what is wrong with them. It also • Transmission of information at all levels of the
helps drivers and open pit mine responsible (s) to have supply chain from top management to implementers.
real time information about vehicles maintenance status. • Analyze data in real time and advice drivers on the
It is the necessary data to make dispatching allocation best course of action
strategies (long term and short term). • Performing self-diagnostics of mining vehicles and
• Performance system: It is the intelligent system that recommending solutions via predictive maintenance.
ensures FMS performance management, through
machine learning models. The datasets are made from This is important because it creates a new era of mining
real and successful open pit mine missions in OCP. This fleet management system in which human errors are reduced
system y is connected to the knowledge base just like all and open pit mine production is optimized in different aspects
the entities of the processing layer. of the transport cycle. This led to better results and cost
Decision layer it is the layer which mainly interfaces the FMS optimization. Figure 3 presents the first version of open pit
with its potential users (directors and executive managers) via: mine platform dashboards. The platform covers ISA95 main
• Collecting the strategic objectives of the open pit mine domains.
from top management and sending them in synchronous
or asynchronous mode to the dispatcher to develop the
mining missions.
• Display of performance indicators and real-time and
instantaneous graphs representing the state of the mine
(production and fleet).
Decision layer is also able to generate stats and administrative
documents in order to facilitate mining regulations respect.
 Be in compliance with standards ✓
Traceability and
and regulations
genealogy
satisfy customer requirements ✓
Real-time analysis of collected ✓
data
Observation SPC (Statistical ✓
Process Control) and SQC
Quality control
(Statistical Control Guide)
Management of non-production ✓
control operations
Initiation of corrective actions ✓
Production control ✓
Control by automatic correction of ✓
Process drifts
management Management of all or part of the ✓
Fig. 3.The first version of open pit mine platform dashboards resources
Alarm management -
Real-time monitoring of indicators ✓
Primary results were positively validated and appreciated by on products and operations
experiment mine actors. Performance
Periodic reports in real time or ✓
management
over the long term
VI. PROPOSED ARCHITECTURE EVALUATION Comparison with the objectives ✓
Checking the conformity of files ✓
Based on the ISA-95 standard[20], which is an Document and forms
international standard for control systems integration, we management Transfer of information between ✓
evaluated the proposed architecture as shown in Table 2. resources
Timely initiation of periodic and ✓
Maintenance preventive interventions
ISA 95 consists of 11 production control functions[21] [18] management History of events ✓
• The control and resource allocation function Diagnostic support tool -
• The dispatching function of production
• The data collection and acquisition function The assessment we conducted to test the maturity of our
• The quality management function architecture with regards to the criteria predefined by ISA 95,
• The management function of the manufacturing process the reference standard in the field, led us to a number of points
• The planning and monitoring function that will be the subject of our future improvements. From a
• The performance analysis function global point of view, our architecture is compliant with the
• The operation management and fine-tuning scheduling various points raised by ISA 95. However, we have concluded
function a number of areas for further improvement, particularly
• The documentation management function concerning alarm management, interoperability with the
• The workforce management function maintenance system and documentation management. As far
as maintenance is concerned, our work will focus on the
• The maintenance management function
development of an automatic communication bridge with the
•
existing maintenance system. The role of this bridge will be to
TABLE II. PROPOSED ARCHITECTURE EVALUATION feed the different models proposed for the proactive
management of dispatching. The second point concerns
Proposed documentation management, as the current architecture does
Function Criteria
architecture
Acquisition from the lowest level ✓ not take into account hard-copy document management, the
Data Acquisistion integration of this module will facilitate the efficient
Automatique collection ✓
Visuel representation of the ✓ implementation of the architecture and its long-term
Dispatching sequencing enhancement.
operations Consideration of unforeseen ✓
circumstances VII. CONCLUSION AND PERSPECTIVES
Real-time staff situation ✓
Monitoring of attendance and ✓ In this article, we proposed an intelligent fleet
Personnel
working time management system focusing on real-time management of
management
Competence and authorisation ✓ mining equipment by applying a multi-agent system. First, we
management
Ressource Know and optimize the mine's ✓
presented a state of the art of fleet management systems as
management resources well as some definitions of the multi-agent system. In
Flow management ✓ addition, we proposed architecture for an intelligent fleet
Information transmitted to the ✓ management system composed of three layers: the vehicle
various workstations layer, the processing layer and the decision layer. We
Product an batch
Consideration of hazards and ✓
tracking implemented the first version of this architecture and test it in
feedback to the dispatcher (rate of
progress, delay, stops) Benguerir experimental mine in order to improve it and extend
Management of outstandings ✓
its application scope. And finally, based on ISA 95 criteria, we [8] L. Zhang and X. Xia, “An Integer Programming Approach for Truck-
Shovel Dispatching Problem in Open-Pit Mines,” Energy Procedia, vol.
evaluated the proposed architecture in view of improving it.
75, pp. 1779–1784, Aug. 2015, doi: 10.1016/j.egypro.2015.07.469.
In perspective, based on our evaluation, we will improve [9] B. Kolonja, D. R. Kalasky, and J. M. Mutmansky, “OPTIMIZATION
dispatching algorithms; we will also detail and implement OF DISPATCHING CRITERIA FOR OPEN-PIT TRUCK HAULAGE
every part of the smart fleet management system architecture. SYSTEM DESIGN USING MULTIPLE COMPARISONS WITH THE
BEST AND COMMON RANDOM NUMBIERS,” p. 9.
[10] “DISPATCH Fleet Management system (FMS) helps mine optimize its
haulage cycle and dramatically reduce truck idle times,” Modular
ACKNOWLEDGEMENT Mining. https://www.modularmining.com/case-studies/dispatch-fms-
helps-mine-optimize-haulage-cycle/ (accessed Nov. 03, 2019).
[11] “Modular Mining | MineSmarterTM,” Modular Mining.
This research was supported by Mohammed VI https://www.modularmining.com/ (accessed Jun. 13, 2019).
Polytechnic University de Benguerir, Morocco and OCP [12] M. Najafi, R. Rafiee, and S. M. E. Jalali, “Open pit limit optimization
group. We thank our colleagues from university and from using dijkstra’s algorithm,” p. 5, 2020.
[13] Shiv PHoomari-Nasaban Ask and Shiv Prakash Upadhyay, “Dynamic
OCP group who provided insight and expertise that greatly shovel allocation approach to short-term production planning in open-
assisted the research. pit mines,” International Journal of Mining, Reclamation and
Environment, 2017.
REFERENCES [14] R. B. Dannenberg, “Time-Flow Concepts and Architectures For Music
and Media Synchronization,” p. 6.
[15] M. Chergui, H. Medromi, and A. Sayouti, “Inter-organizational
[1] S. Alarie and M. Gamache, “Overview of Solution Strategies Used in Workflow for Intelligent Audit of Information Technologies in terms of
Truck Dispatching Systems for Open Pit Mines,” Int. J. Surf. Min. Entreprise Business Processes,” Int. J. Adv. Comput. Sci. Appl., vol. 5,
Reclam. Environ., vol. 16, no. 1, pp. 59–76, Mar. 2002, doi: no. 5, 2014, doi: 10.14569/IJACSA.2014.050515.
10.1076/ijsm.16.1.59.3408. [16] A. M. Nascimento et al., “A Systematic Literature Review About the
[2] J. A. Curry, M. J. L. Ismay, and G. J. Jameson, “Mine operating costs Impact of Artificial Intelligence on Autonomous Vehicle Safety,” IEEE
and the potential impacts of energy and grinding,” Miner. Eng., vol. 56, Trans. Intell. Transp. Syst., pp. 1–19, 2019, doi:
pp. 70–80, Feb. 2014, doi: 10.1016/j.mineng.2013.10.020. 10.1109/TITS.2019.2949915.
[3] A. M. Afrapoli and H. Askari-Nasab, “Mining fleet management [17] R. Salazar-Cabrera, A. P. De La Cruz, and J. M. Madrid Molina, “Fleet
systems: a review of models and algorithms,” Int. J. Min. Reclam. Management and Control System from Intelligent Transportation
Environ., vol. 33, no. 1, pp. 42–60, Jan. 2019, doi: Systems perspective,” in 2019 2nd Latin American Conference on
10.1080/17480930.2017.1336607. Intelligent Transportation Systems (ITS LATAM), Bogota, Colombia,
[4] F. M. Frédéric Semet, “Les Problèmes De Gestion De Flotte En Temps Mar. 2019, pp. 1–7, doi: 10.1109/ITSLATAM.2019.8721347.
Réel,” INFOR Inf. Syst. Oper. Res., vol. 44, no. 4, pp. 299–330, Dec. [18] V. Zeimpekis, G. M. Giaglis, and I. Minis, “Development and
2006, doi: 10.1080/03155986.2006.11732754. Evaluation of an Intelligent Fleet Management System for City
[5] W. B. Powell, P. Jaillet, and A. Odoni, “Chapter 3 Stochastic and Logistics,” in Proceedings of the 41st Annual Hawaii International
dynamic networks and routing,” in Handbooks in Operations Research Conference on System Sciences (HICSS 2008), Waikoloa, HI, USA, Jan.
and Management Science, vol. 8, Elsevier, 1995, pp. 141–295. 2008, pp. 72–72, doi: 10.1109/HICSS.2008.121.
[6] A. Moradi Afrapoli, M. Tabesh, and H. Askari-Nasab, “A multiple [19] H. Billhardt et al., “Dynamic Coordination in Fleet Management
objective transportation problem approach to dynamic truck dispatching Systems: Toward Smart Cyber Fleets,” IEEE Intell. Syst., vol. 29, no. 3,
in surface mines,” Eur. J. Oper. Res., vol. 276, no. 1, pp. 331–342, Jul. pp. 70–76, May 2014, doi: 10.1109/MIS.2014.41.
2019, doi: 10.1016/j.ejor.2019.01.008. [20] L. Yue, P. Niu, and Y. Wang, “Guidelines for defining user requirement
[7] D. K. Ahangaran, A. B. Yasrebi, A. Wetherelt, and P. Foster, “Real – specifications (URS) of manufacturing execution system (MES) based
time dispatching modelling for trucks with different capacities in open on ISA-95 standard,” J. Phys. Conf. Ser., vol. 1168, p. 032065, Feb.
pit mines / Modelowanie w czasie rzeczywistym przewozów ciężarówek 2019, doi: 10.1088/1742-6596/1168/3/032065.
o różnej ładowności w kopalni odkrywkowej,” Arch. Min. Sci., vol. 57, [21] “Le standard ISA 95 - Smart Manufacturing,” TLG Pro, Feb. 05, 2018.
no. 1, pp. 39–52, Oct. 2012, doi: 10.2478/v10267-012-0003-8. https://www.tlgpro.fr/2018/02/standard-isa-95/ (accessed Jan. 02, 2019).