1th International Conference

„CONFERENCE ON ADVANCES IN SCIENCE AND TECHNOLOGY“ COAST 2022

May 26-29, 2022 HERCEG NOVI, MONTENEGRO

NATURE-INSPIRED OPTIMIZATION ALGORITHMS FOR

SUPPLY CHAIN MANAGEMENT PROBLEM: A REVIEW
Olga Ristić1, Sandra Milunović Koprivica1
1
Faculty of Technical Sciences, University of Kragujevac, 32000 Čačak, Republic of
Serbia

Corresponding author e-mail address: [email protected] (O. Ristić)

-----------------------------------------------------------------------------------------------------------
ABSTRACT:
In this paper will be analyzing the modern age nature-inspered optimization algorithms
which are used in solving supply chain management (SCM) problems. The main aim of
this paper is to give a review of research related to implementation of optimization
algorithms in all domain of SCM. Here is describe the SCM with proposed domains and
subdomains. References published in the last few years related to the optimization
problems in various areas of SCM are presented. Finally, an analysis of optimization
algorithm which were used for SCM are discussed. The impacts of algorithm
modification and hybridization as well as future research opportunities are analyzed.

Keywords: supply chain, optimization algorithm, TSP,

-----------------------------------------------------------------------------------------------------------------------

1. INTRODUCTION

Nature-inspired optimization algorithms (NIOAs), defined as a group of algorithms that

are inspired by natural phenomena, including swarm intelligence, biological systems,
physical and chemical systems, human inspired optimization algorithms etc. NIOAs
have made significant progress within the last 30 years. Thus far, a large number of
NIOAs implemented in SCM processes such like Genetic Algorithm (GA), Particle
Swarm Optimization (PSO) algorithm, Artificial Bee Colony (ABC) algorithm, Cuckoo
Search (CS) algorithm, Ant Colony Optimization (ACO) algorithm, Bat Algorithm
(BA), Differential Evolution (DE) algorithm, Firefly Algorithm (FA), Grey Wolf
Optimization (GWO), Emperor Penguins Colony (EPC) algorithm, Gravitational Search
Algorithm (GSA), Teaching-Learning Based Optimization (TBLO) algorithm, Harmony
Search (HS) algorithm and etc. [1]. All of those algorithms have been applied to hard
optimization problems. Their main characteristic is the imitation of the way natural
systems function and evolve in order to deal with real-world situations [2]. For example,
ant colonies cooperate to find high-quality food source, a swarm of birds implements a
scheme of indirect communication with the aim of finding the optimal direction, etc. In
addition to the theoretical studies of NIOAs, we summarize the current survey work of
the NIOAs, discuss our motivations for this survey, and scope of this work.

SCM is one of the most important areas in the industry and services. It is responsible for
procurement, storage and maintaining of raw materials, inventory management and
 1th International Conference
„CONTEMPORARY ADVANCEMENT IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

transportation final products from production to end-users (customers) locations. The

goal of SCM is to use resources in order to produce goods which will have the best
quality with optimization of time and cost issues [3]. A supply chain is consists of
independent organizations such like suppliers, logistics providers, manufacturers, and
distributors who all work in an integrated system to add value to a product. Applying the
different optimization algorithms may provide the manufacturers with advantages such
as less transportation costs, the possibility of using the cheap workforce or materials,
lower production time, and reducing the risk of production due to natural disasters. The
list of activities of producing the sufficient quantity of useful product at the right time is
classified to the supply chain problem [4].

2. SUPPLY CHAIN MANAGEMENT

Supply chain network (SCN) is basically a combination of many components. They are
suppliers, manufacturing plants, distribution centers, retailers and customers as it
presented on Fig. 1 [5]. It is also the collection of physical locations of manufacturing
and distribution center, transportation vehicles and all supporting systems through which
the products and services company markets are managed and finally delivered.

The effectiveness and usability of the system are mainly affected by some key factors.
The major factor is to search the best transportation scheduling for any SCN to minimize
the cost for transporting row materials from suppliers to manufacturing plants and
products from manufacturing to the end customer [6]. Many researchers studies,
approaches and methodologies to solve the different problems of SCM by NIOAs
approaches.

SCM covers a variety of NP-hard optimization problems such as travelling salesman

problem (TSP), vehicle routing problem (VRP), knapsack problem, team optimization,
inventory, supply network problems, etc. NIOAs are capable of determine near optimal
solutions for instances of those problems in a reasonable amount of time.

Fig. 1. Example of SCN

 1th International Conference
„CONFERENCE ON ADVANCES IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

In order to explain the basics of SCM, it is usfull to present the main functions of a
simplified supply chain [7]. The modelling approach consists of three main parts: the
logistic, the supplying and the distribution system.

Logistic system receives orders from customers and dispatch orders to the suppliers. The
purchased row materials are going to be used in the manufacturing process. In this phase,
the main objective is to minimize the tardiness time. Tardiness time is the difference
between the completion date and the desired delivery date of the final product. However,
there are two disorder. Firstly, suppliers will not be able to deliver raw materials on time
due to some constraints. Secondly, customers might ask for delivery dates which are not
matching with the suppliers’ services. The logistic problem are often formulated by
using the overall assignment optimization approach.

Supplying system consists of a network of various suppliers or manufacturers and each

one participate in producing the requested product components for the logistic system.
The minimization of the total tardiness time, as defined above can be considered as the
objective function. The supply problem can be considered as a scheduling optimization
problem.

Distribution system delivers the final products to the corresponding clients. Clients
might be classify by their geographical location. A simple real model of the distribution
problem is the vehicle routing problem (VRP) or the travelling salesman problem (TSP).
This is a problem which considers the transportation low cost of the fleet of available
vehicles. For this type of problem, a typical constraint is that every customer is visited
by one vehicle, and all customers should to be satisfied.
All activities carried out within the SCM can be classified into domains and subdomains
as shown in Table 1. In this way, it is possible to pay special attention to critical domains
that affect the time of production and delivery, total costs and human resources in the
delivery of final product.

Table 1. Proposed domains and subdomains in SCM

Domains Subdomains
D1.1-Supplier Selection
D1-Procurement D1.2-Supplier Evaluation
D1.3-Inventory Investment Management
D1.4-Standardization
D1.5-Make or Buy Analysis
D1.6-Purchasing Decisions
D1.7-Quote Management
D2.1-Order Processing
D2.2-Order Confirmation and Tracking
D2-Purchasing D2.3-Order Completion approval
D2.4-Documentary Management
D2.5-Purchasing Efficiency Management
D3-Inventory D3.1-Risk management
Management D3.2-Order Decision
 1th International Conference
„CONTEMPORARY ADVANCEMENT IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO
D3.3-Cost management
D3.4-Decision about minimum and optimal inventory
D4.1-Mode, Vehicle, Route Selection
D4.2-Vehicle Loading
D4- D4.3-Vehicle Tracking, Vehicle Condition Minitoring
Transportation D4.4-Document and Regulatory Management
D4.5-Transportation Cost Management
D4.6-Own or external process
D5.1- Forecasting Data Management and Integration Management
D5-Demand
D5.2- Forecasting Techniques
Forecasting
D5.3- Forecasting Efficiency
D6.1-Inventory Quntity/Quality Tracking
D6.2-Movement Management/Technology
D6-Wharehousing D6.3-Warehouse Capacity/Condition Tracking
D6.4-Warehouse Efficiency Management
D6.5-Own or external process
D7.1-Labeling Management
D7.2-Handeling Management
D7-Packaging
D7.3- Apportionment Management and Packing Efficiency Management
D7.4- Own or external process
D8.1-Movement Tracking
D8-Inner
D8.2-Movement Technology Management
Distribution
D8.3-Movement Cost/Efficienty Management
D9.1- Handling Standardization and Handling Information Management
D9-Handling
D9.2- Handling Technology Management
D10.1-Maintenance Planning Management and Asset Condition Tracking
D10.2-Predictive Maintenance Management
D10.3-Maintenance Efficiency Management
D10-Maintenance
D10.4-Spare Part Inventory Management
D10.5-Maintenance Technology Management
D10.6-Own or external process
D11.1-Written Statement of Policy, Customer Receipt of Policy
Statement
D11.2-Organization Structure
D11.3-System Flexibility
D11.4-Management Services
D11-Customer D11.5-Stockout Lavel
Service D11.6-Order Information dan Elements of Order Cycle
D11.7-Expedited Shipments and Transshipments
D11.8-System Accuracy and Order Convenience
D11.9-Installation, Warranty, Alternations, Repairs and Parts
D11.10-Product Tracing
D11.11-Customer Claims, Complaints and Returns

3. ROF PERFORMANCE OF SUPPLY CHAIN

Measurement system in supply chains must focus on measuring three different types
performance: resource measurement (R), output measurement and (O) and flexibility
measurement (F) [8]. All three types performances have different goals, as it is shown in
 1th International Conference
„CONFERENCE ON ADVANCES IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

Table 2. A measurement must be performed all three performances because each of them
contributes supply chain performance and has an impact on other characteristics.

Resource reduction can negatively affect on flexibility in supply. Supply system can
currently use such reduced resources, and produce the desired outputs, but will such a
chain be able to adapt to change (eg. product requirements, unreliability production,
when introducing new products, or reduction of bidders). Accordingly, flexibility is very
important to consider in the supply chain performance.

Table 2. Supply chain performance goals

Important
Performance type Goals Purpose / Significance
performances
High level Resources efficiency Total costs
efficiency have crucial influence on Distribution costs
profitability Production costs
Resources (R)
Stock status
ROI - return to
investment
High level Without acceptable Total income
services to the output, users will turn to Profit
user others suppliers Satisfaction rate
Delivery accuracy
Output (O) Arrears / deficit
Response time to the user
Preparation time of production
Delivery errors
User complaints
Ability reactions In uncertain Range of flexibility
on change in the environment, supply Sensitivity of flexibility
Flexibility (F)
surroundings chains must be able to
answer to the changes

4. NATURE INSPIRED OPTIMIZATION ALGORITHMS IMPLEMENTED IN

SCM

In the last 30 years, there has been developed many NIOAs for solving different
problems. This algorithms come to be widespread because of their ability to modify to
any changing environment especially for optimization. Many researchers has inspired by
the nature in solve optimization problems in different areas. Here, we make review of
papers used NIOAs to solve complex SCM problems. Hybrid algorithms created by
combining two or more NIOAs are used to solve some complex problems more
efficiently [8]. Different NIOAs showed high-performance in solving different SPM
problems as summarized in Table 3.

Table 3 shows only few references that are used in solving SCM-related optimization
problems. The GA is the most commonly used algorithm, and it became popular in the
1970s. The other optimization algorithms were developed lately. Most of them are based
on algorithms that were inspired by some animals and their behavior or hunting in
nature. This has led to the development of a large number of different algorithms that
efficiently optimize some important parameters in SCM such as minimizing production
 1th International Conference
„CONTEMPORARY ADVANCEMENT IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

time, supply time, final product delivery or profit maximization. When solving various
problems that arise, special attention should be paid to the existing production capacities
(production line, machine, worker), the number of transport vehicles, the size of the
warehouse, etc. The application of the algorithm for solving that problem will also
depend on the complexity of the analyzed system.

Table 3. Summarize NIOAs used in supply chain published papers

Algorithms
No Application Discussion
& Reference
In this paper, the joint replenishment problem is modeled for a TLBO, GA
two-level supply chain consisting of a single supplier and [9]
Procurement

multiple retailers that use the for specific products. The main GWO
D1-

1. goal is to find the optimal number of products. This is [10]

acomplish by using Teaching-Learning Based Optimization
Algorithm-TLBO and checked with Genetic Algorithm. GWO
algorithm is used for efficient selection of suplier.
In this paper, a modified version of the GWO scheme is GWO
introducedto solve the complex optimization problems of [11]
supplier selection and order quantity allocation. The improved
D2-Purchasing

GWO method called iGWO includes weighted factors and a

displacement vector to promote the exploration of the search
2. strategy, avoiding the use of inconviniant solutions. The
proposed algorithm has been tested on a different number of
instances of a difcult problem. The results show that the
proposed algorithm obtains the optimal cost solutions,
maintains a better search strategy, finding feasible solutions in
all instances.
The purpose of this study is to give a new dimension on GA, PSO
warehouse with PSO using GA in supply chain in inventory [12]
D3-Inventory
Management

optimization and to develop more realistic and flexible

models. GA is used for improve the effectiveness and
3.
efficiency for various aspect of warehouse. The results of the
proposed model help researchers to think about optimizing
their complex problems using different evolutionary search
algorithm.
Here is studied and analyzed the nature of transportation CRO
vehicle scheduling problem in a SCN with the help of logistics algorithm
enterprise by using a meta-heuristic algorithm called chemical [13]
D4-Transportation

reaction optimization (CRO). All the transportation nodes GA

classified into three distinct classifications. It is proposed a [14]
CRO algorithm using four reaction operators with an
4.
additional repair operator to find out the best transportation
routes within shortest computing time.
For minimization of transportation cost is also used GA. The
proposed model is helpfulfor fresh food quality control and
fresh food supply chain to make a strategic decision about
transportation expenses
In this paper is presented a novel method of PSO. Here is used PSO
D5-Demand
Forecasting

the retail sales data based on the monthly sales report for retail [15]
trade and food services report to test the accuracy for different
5. values of accommodation coefficient. It is achieved lower
error rates in forecasts which results in better sales predictions.
 1th International Conference
„CONFERENCE ON ADVANCES IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO
The main contribution of using the EPC algorithm in the field EPC, GA,
of inventory control to reduce computational cost and its PSO
D6-Wharehousing comparison with the other NIOAs. The main goal was to [16,17]
achieve the best fitness function result with the lowest cost and
runtime. Experiments were performed for each model with
6.
two different population sizes. In the deterministic single-
product model, the EPC algorithm with different population
sizes performed best. PSO and IWO algorithms also provided
acceptable solutions. But an algorithm like GA provided
mediocre performance.
Here is developing a new evolutionary heuristic approach for MGA, TS,
solving the three-dimensional packing problem called BPP SA, ACO
D7-Packaging

(Bin packing problem). It is applied to the variation of [18]

problem with a single container and a set of boxes of various
7.
dimensions, called the SKP (Single knapsack problem) using
modified genetic algorithm (MGA), TS, SA, ACO algorithms.
Thus, it was confirmed that MGA and NIOAs can be
effectively used to solve many logistic NP-difficult problems.
The significant algorithm described in paper has been ranked GA, PSO,
into five primary categories: ACO, PSO, GA, ABC, and ACO, ABC
D8-Inner Distribution

memetic algorithm. They have been applied for optimization [19]

and design of logistics distribution systems. It can leading to CSO
the elevation of the effectiveness of logistics enterprises. Also, [20]
8. the policy assistance for the logistics organization has been
empowered to boost the efficient and healthy progress of it.
GA and its hybrids have display the best efficiency until now.
It is essential to investigate the logistics distribution route
optimization by NIOAs for optimizing the logistics and
selecting a rational distribution scheme.
The supply chain finance industry will generate the flow of GA
funds and commodities when providing financing services to [21]
small and medium-sized enterprises (SMEs). The investigation
on supply chain finance under information sharing from the
aspect of credit risk assessment will be conducted. The GA is
D9-Handling

selected to evaluate the credit risk of supply chain finance. In

9. addition, among all the investigated enterprises, the cost gap is
large, and the standard deviation of the inventory decline price
reserve is small, which means that most enterprises have good
inventory quality. The supply chain financial risk assessment
is mainly explored from the perspective of banks. The results
can provide theoretical support for reducing the probability of
bank’s profit damage and increasing the bank’s profitability.
The quantum GA is used to optimize the predicted health GA
Maintenance

status of the equipment. The use of quantum GA to optimize [22]

D10-

the inspection path compared with the traditional GA, the

10. objective function value is reduced by about 50%. thus saving
inspection time.

A cost optimization problem is developed to obtain the BA, PSO

optimal values of system design parameters. The comparative [23]
D11-Customer

analysis of proposed bat algorithm (BA) with particle swarm

Service

optimization (PSO) to achieve the optimal operating

11.
conditions with minimal expected cost. Numerical simulations
and illustrations are provided to get an understanding of the
mathematical modeling in detail and performance analysis of
the queueing measures.
 1th International Conference
„CONTEMPORARY ADVANCEMENT IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

5. CONCLUSION

There are many SCM problem, which can be solved by proposed NIOAs. Many
scientists try to give mathematical model for reducing total costs, production time,
delivery time, flexible responding to customer demands, limited storage capacity,
making higher profits to find best solution for effective SCM [24]. GA as well as hybrid
models are most commonly used to solve supply chain optimization problems. The
authors came to various conclusions in the application of different algorithms in order to
achieve the best possible results.
Future research should include the importance of a sustainable supply chain, as well as
factors influence the environmental protection. Since a large number of production
systems, as well as transportation systems have a negative impact on pollution due to gas
emissions, special attention should be paid to these factors as well.

Acknowledgement

This study was supported by the Ministry of Education, Science and Technological
Development of the Republic of Serbia, and these results are part of the Grant No. 451-
03-68/2022-14/200132 with the University of Kragujevac-Faculty of Technical Sciences
Čačak.

6. LITERATURE

[1] Mirjalili, S.M., Dong, J.S., & Lewis, A. (2020). „Nature-Inspired Optimizers -
Theories, Literature Reviews and Applications“. Nature-Inspired Optimizers, 239.
[2] Xu, X.; Hao, J.; Zheng, Y. (2020). „Multi-objective artificial bee colony algorithm
for multi-stage resource leveling problem in sharing logistics network“,
Computers & Industrial Engineering, Vol. 142, 106338.
[3] Matinrad, N., Roghanian, E. & Rasi, Z. (2013). „Supply chain network
optimization: A review of classification, models, solution techniques and future
research“, Uncertain Supply Chain Management, Vol. 1, 1–24.
[4] Rajyashree, R., Pathak, P., Upadhayay, S., & Garg, V. (2018), „Big Data Analysis
on Supply Chain Management“, 3rd International Conference on Communication
and Electronics Systems (ICCES), 1145-1150.
[5] Beheshtinia, M.A. Feizollahy, P. Fathi, M. (2021). „Supply Chain Optimization
Considering Sustainability Aspects“. Sustainability, 13, 11873.
[6] Dolgui, A., Ivanov, D., Sethi, S., & Sokolov B., (2019), „Scheduling in
production, supply chain and Industry 4.0 systems by optimal control:
fundamentals, state-of-the-art and applications“, International Journal of
Production Research, Taylor & Francis, Vol. 57, No. 2, 411-432.
[7] Soni, G., Jain, V., Chan, F. T. S., Niu, B., & Prakash, S. (2018). „Swarm
intelligence approaches in supply chain management: potentials, challenges and
future research directions“, Supply Chain Management: An International
Journal,Vol. 24, No. 1, 107-123.
 1th International Conference
„CONFERENCE ON ADVANCES IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

[8] Jagan Mohan Reddy K., Neelakanteswara Rao A., Krishnanand L. (2019), „A
review on supply chain performance measurement systems“, Procedia
Manufacturing, Nо. 30, 40-47.
[9] Slowik, A., & Cpalka, K. (2021). „Hybrid Approaches to Nature-inspired
Population-based Intelligent Optimization for Industrial Applications“. IEEE
Transactions on Industrial Informatics, Vol. 18, No. 1, pp. 546-558.
[10] Hamid Reza Pasandideh, S., Taghi Akhavan Niaki, S., Ahmadi, P. (2018).
"Vendor managed inventory in the joint replenishment problem of a multi-product
single-supplier multiple retailer supply chain: A teacher-learner-based
optimization algorithm", Journal of Modelling in Management, Vol. 13 Issue: 1,
156-178.
[11] Alejo-Reyes, A., Cuevas, E., Rodríguez, A., Mendoza, A., & Olivares-Benitez, E.
(2020). „An Improved Grey Wolf Optimizer for a Supplier Selection and Order
Quantity Allocation Problem“, Mathematics, Vol. 8, 1457;.
[12] Talati, I., Mishra, P., Shaikh, A. (2021). „An Analytic and Genetic Algorithm
Approach to Optimize Integrated Production-Inventory Model Under Time-
Varying Demand“. In: Shah, N.H., Mittal, M. (eds) Soft Computing in Inventory
Management. Inventory Optimization.
[13] Islam, M.R., Mahmud, M.R., & Pritom, R.M. (2019). „Transportation scheduling
optimization by a collaborative strategy in supply chain management with TPL
using chemical reaction optimization“. Neural Computing and Applications, 32,
3649-3674.
[14] Hongli, Y., & Yongming, W. (2017). „Transportation expenses minimal modelling
with application to fresh food supply chain“. 13th IEEE International Conference
on Electronic Measurement & Instruments (ICEMI), 263-267.
[15] Mittal, M., Swain, P.R., & Rana, H. (2019). „A Nature Inspired Optimisation
Method for Supply Chain Management Problem“. 2019 Amity International
Conference on Artificial Intelligence (AICAI), 505-509.
[16] Harifi, S., Khalilian, M., Mohammadzadeh, J., & Ebrahimnejad, S. (2021).
„Optimization in solving inventory control problem using nature inspired Emperor
Penguins Colony algorithm“. Journal of Intelligent Manufacturing, Vol. 32, 1361-
1375.
[17] Yadav, A.S., Maheshwari, P., & Swami, A. (2016). „Analysis of Genetic
Algorithm and Particle Swarm Optimization for Warehouse with Supply Chain
Management in Inventory Control“. International Journal of Computer
Applications, 154, 10-17.
[18] Mokshin, V. V., Maryashina, D., Stadnik, N., Zolotukhin, A. and Sharnin, L.
(2020). “Modified genetic algorithm as a new approach for solving the problem of
3d packaging.” 6th International Conference Information Technology and
Nanotechnology, ITNT-DS 2020, Samara, May 26–29, 91-97.
[19] 12 Wang, C., Qian, Y. & Shaic, S. (2021). „The Applications of Nature-Inspired
Algorithms in Logistic Domains: A Comprehensive and Systematic
Review“, Arabian Journal for Science and Engineering, Vol. 46, 3443–3464.
[20] Manimaran, P., & Selladurai, V. (2014). „Cat swarm optimization for single stage
supply chain distribution system with fixed charges“, ICTACT Journal on Soft
Computing, Vol. 04, No. 02, 687-691.
 1th International Conference
„CONTEMPORARY ADVANCEMENT IN SCIENCE AND TECHNOLOGY“ COAST 2022
May 26-29, 2022 HERCEG NOVI, MONTENEGRO

[21] Sang, B. (2021). „Application of genetic algorithm and BP neural network in

supply chain finance under information sharing“. Journal of Computational and
Applied Mathematics, Vol. 384, 113170.
[22] Kang, W., Jin, W., Jiang, P., & Shi, L. (2020). „Distribution Network Equipment
Fault Prediction and Maintenance Path Optimization Strategy Based on Deep
Learning and Knowledge Map Comprehensive Analysis“. 2020 2nd International
Conference on Artificial Intelligence and Advanced Manufacture (AIAM). 42-49.
[23] Shekhar, C., Varshney, S., & Kumar, A. (2020). „Optimal control of a service
system with emergency vacation using bat algorithm“. Journal of Computational
and Applied Mathematics, Vol. 364, 112332.
[24] Wang, C., Qian, Y. & Shaic, S. (2021), The Applications of Nature-Inspired
Algorithms in Logistic Domains: A Comprehensive and Systematic
Review. Arabian Journal of Science and Engineering, No. 46, 3443–3464.