2018Unpublished venueOpen access

Modelling sustainable supply networks with adaptive agents

Subodha Dharmapriya, Senevi Kiridena, Nagesh Shukla

Open full text 1 citations

Abstract

This paper proposes a multi-agent modelling approach that supports supply network configuration decisions towards sustaining operations excellence in terms of economic, business continuity and environmental performance. Two types of agents are employed, namely, physical agents to represent supply entities and auxiliary agents to deal with supply network configuration decisions. While using the evolutionary algorithm, Non-dominated Sorting Genetic Algorithm-II to optimize both cost and lead time at the supply network level, agents are modelled with an architecture which consists of decision-making, learning and communication modules. The physical agents make decisions considering varying situations to suit specific product-market profiles thereby generating alternative supply network configurations. These supply network configurations are then evaluated against a set of performance metrics, including the energy consumption of the supply chain processes concerned and the transportation distances between supply entities. Simulation results generated through the application of this approach to a refrigerator production network show that the selected supply network configurations are capable of meeting intended sustainable goals while catering to the respective product-market profiles.

About this research paper

What this paper is about

This paper proposes a multi-agent modelling approach that supports supply network configuration decisions towards sustaining operations excellence in terms of economic, business continuity and environmental performance. Two types of agents are employed, namely, physical agents to represent supply entities and auxiliary agents to deal with supply network configuration decisions. While using the evolutionary algorithm, Non-dominated Sorting Genetic Algorithm-II to optimize both cost and lead time at the supply network level, agents are modelled with an architecture which consists of decision-making, learning and communication modules. The physical agents make decisions considering varying situations to suit specific product-market profiles thereby generating alternative supply network configurations. These supply network configurations are then evaluated against a set of performance metrics, including the energy consumption of the supply chain processes concerned and the transportation distances between supply entities. Simulation results generated through the application of this approach to a refrigerator production network show that the selected supply network configurations are capable of meeting intended sustainable goals while catering to the respective product-market profiles.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper proposes a multi-agent modelling approach that supports supply network configuration decisions towards sustaining operations excellence in terms of economic, business continuity and environmental performance. Two types of agents are employed, namely, physical agents to represent supply entities and auxiliary agents to deal with supply network configuration decisions. While using the evolutionary algorithm, Non-dominated Sorting Genetic Algorithm-II to optimize both cost and lead time at the supply network level, agents are modelled with an architecture which consists of decision-making, learning and communication modules. The physical agents make decisions considering varying situations to suit specific product-market profiles thereby generating alternative supply network configurations. These supply network configurations are then evaluated against a set of performance metrics, including the energy consumption of the supply chain processes concerned and the transportation distances between supply entities. Simulation results generated through the application of this approach to a refrigerator production network show that the selected supply network configurations are capable of meeting intended sustainable goals while catering to the respective product-market profiles.

Key concepts: Supply chain, Supply chain network, Supply network, Computer science, Sorting, Genetic algorithm, Set (abstract data type), Product (mathematics)

Related papers

Back to paper searchBrowse research topicsOriginal source
Modelling sustainable supply networks with adaptive agents — Research Paper | ScholarLens