Production planning and control system on the factory floor using the cooperation of intelligent agents to achieve mass customization

Introduction

Recent advances in flexible manufacturing and information technology have made it possible for production systems to provide a wider range of products or services at a lower cost. In addition, the increase in competition at the global level has led to the confrontation of industries with the approach of increasing customer value in providing products or services. Therefore, the need to consider the specific needs of each customer has led manufacturers to involve customers in the production process. In the meantime, customization in mass production [1] is one of the new production methods that is being noticed by more producers every day. Customization in mass production is the ability to produce a product or service specific to each customer based on their order or known needs through a fully flexible and integrated process while maintaining the benefits of mass production. Customization in mass production is one of the opportunities provided to producers through the growth and coordination of production technologies and information technology.

It is clear that the activities carried out in customization systems in mass production require extensive cooperation, information exchange and interaction within and outside the organization. Part of this interaction is to plan the assignment of tasks to resources and schedule the execution of tasks on resources. The problem of planning the assignment of tasks to resources and scheduling the execution of tasks is considered one of the most complex problems of optimization of compositions, which in this research is an attempt to achieve an expansion in the field of solving this category of problems.

In customization systems in mass production, since each manufactured item will have its own conditions, based on the stated needs of the customer, the problem of coordination and interaction of components in the factory takes on a more complex form. In order to solve the planning problem in such conditions, the components of the planning system must have characteristics such as communication with other components, responsiveness and autonomy. According to these characteristics, the use of multi-agent evolutionary systems is proposed as one of the appropriate solutions. In this method, by taking advantage of the society of agents, each of which has a set of characteristics and resources, solutions can be obtained in a dynamic computing environment. flexible, a new and effective model for modeling the solution space of the problem is proposed. The presented model has special characteristics, among the most important of which we can point out the simultaneous solution of the sub-problems of planning the assignment of tasks to resources and the scheduling of the order of execution of tasks. In the continuation of the research, the optimization algorithm of the collective movement of particles is introduced and a new form of this algorithm is introduced to solve multi-objective optimization problems, in which the selection of guide particles is made based on the density of particles in the target space, then the presented algorithm is compared with one of the similar algorithms. After that, the two presented ideas will be used in solving the problem of flexible workshop work scheduling and its results will be examined. They are examined from the point of view of different authors and researchers, and the problem of planning and scheduling production to achieve customization in mass production is reduced to the problem of planning and scheduling a flexible production workshop [2] with several goals [3]. From here on, the research approach will be towards solving the multi-objective scheduling problem in flexible workshop work. In the third chapter, the concepts of workshop scheduling are presented and by using them, a suitable mathematical model is obtained for the problem of flexible workshop work scheduling. In the continuation of the thesis, in the fourth chapter, a short review of the concept of metaexploration [4] and its application in optimization problems is done.. In this chapter, some important meta-exploratory methods that are widely used in optimization are introduced in the form of three categories. In the fifth chapter, multi-objective optimization methods are studied based on the optimization algorithm of collective motion of particles [5] as an evolutionary [6] multi-factor [7] method, and a new algorithm based on the kernel density [8] of particles in the target space is presented and compared with one of the multi-objective optimization algorithms based on collective motion of particles. In the sixth chapter, first, a new representation of the search space for the flexible workshop work scheduling problem is introduced, which provides the ability to simultaneously solve the sub-problems of assigning operations to machines and scheduling the order of execution of operations simultaneously. Then, in the continuation of this chapter, the optimization algorithm that was introduced in the fifth chapter is used to solve the flexible workshop work scheduling problem and its results are compared with other methods. Finally, in the seventh chapter, the discussion and conclusions about the results of the research conducted in this thesis are discussed and suggestions for the expansion of research in this field are presented.