Highlights

Abstract

Keywords

۱٫ Introduction

۲٫ Related work

۳٫ Construction of conceptual model

۴٫ Transformation from conceptual model to simulation model

۵٫ Application: A case study

۶٫ Conclusion and future work

CRediT authorship contribution statement

Declaration of Competing Interest

Acknowledgement

References

Abstract

Agent-based modeling is a useful and effective approach to exploring industrial logistics system, while agent interaction protocol is a key to support agent-based industrial logistics modeling. Current modeling methods are mainly developed to face general industrial systems and are in a certain sense difficult to fully meet the need of modeling military industrial logistics system.

In this research, we focus on describing and implementing military industrial logistics agent interaction protocol by presenting a new command interaction diagram (CID) oriented paradigm. We not only present the concept and corresponding graphical notation representations of CID for multi-agent interaction process, but also design and develop the CID modeling tool with a function of logic verification, thus realizing the mechanism of automatic transformation from conceptual model of CID to its simulation model. In addition, we provide a case study and perform agent-based military industrial logistics simulation. The simulation demonstration and corresponding validation show that our method can meet actual needs and achieve good results, with more advantages than other traditional modeling methods. This fact proves the feasibility and effectiveness of the proposed method.

۱٫ Introduction

Industrial logistics plays a key role in industrial engineering (Li et al., 2019). It involves the physical inflow and outflow of goods and corresponding services (Barros, 1997). Of course, it can be further divided into product support logistics, production support logistics and industry support logistics (Barros, Riley, & Brown, 2001; Li et al., 2019). With the evolution of technology, the methods for warfare have also changed. In modern wars, as the military application field of industrial logistics, military industrial logistics has an obvious characteristic in logistics entity interaction based on both command orders and information sharing, thus forming an integrated system support capability. In view of this new support mode, it is necessary to conduct more scientific and reasonable analysis and research by establishing simulation models of military industrial logistics system. Modeling military industrial logistics entity interaction has become a key bottleneck problem in exploring the running mechanism of military industrial logistics system. Although some methods, such as exercise and training, man-in-theloop simulation experiment, can be used in simulation for military industrial logistics system, there are some disadvantages, including abundant capitals, high cost and inconvenience in operation. As far as traditional analytical modeling methods are concerned, generally, they are difficult to describe the intelligent behavior characteristics of military industrial logistics system. Intelligent agent is a program that maps percepts to actions. It acquires information from its environment (“perceives” the environment) and decides about its actions and performs them (Lesser, 1998; Wooldridge & Jennings, 1999; Shi, 2001). Obviously, the internal member entities of military industrial logistics system, such as logistics command vehicles, repair vehicles and rescue vehicles, can be viewed as intelligent agents. Thus, using agent-based modeling method to describe military industrial logistics entities is a feasible and effective solution.