MILP Modeling and Optimization of Energy- Efficient Distributed Flexible Job Shop Scheduling Problem

Abstract

With the global warming problem and increasing energy cost, manufacturing firms are paying more and more attention to reducing energy consumption. This paper addresses the distributed flexible job shop scheduling problem (DFJSP) with minimizing energy consumption. To solve the problem, firstly, a novel mixed integer linear programming (MILP) model is developed to solve small-scaled problems to optimality. Due to the NP-hardness of DFJSP, we then propose an efficient hybrid shuffled frog-leaping algorithm (HSFLA) for solving DFJSP, particularly for large-sized problems. HSFLA combines the shuffled frog-leaping algorithm (SFLA) with powerful global search ability and variable neighborhood search (VNS) with good local search ability. Moreover, in HSFLA, the encoding method, the decoding method, the initialization method and the memetic evolution process are specifically designed. Finally, numerical experiments are conducted to evaluate the performance of the proposed MILP model and HFSLA. INDEX TERMS Distributed flexible job shop scheduling problem, mixed integer linear programming, shuffled frog-leaping algorithm, variable neighborhood search, energy consumption.

Federated Deep Reinforcement Learning for Prediction-Based Network Slice Mobility in 6 G Mobile Networks

“…Here, (9b) and (9c) ensure that the allocated resources of VNFs in the physical host will not exceed the total resources of the host; (9d) ensures that the serving latency of users in slices meet the SLA requirements; (9e) ensures that each VNF selects only one host for migration and has only two distinct migration statuses at each time slot; (9f) ensures that each VNF can only be assigned to one slice; (9g) ensures a thorough weight setting for measuring the prices of resources. The above-formulated problem is a Mixed Integer Linear Programming (MILP) problem due to the integer variables Θ v,X (t), ∆ v (t), π v (t) and the continuous variable R v,i (t), which is proved to be NP-hard [42], we consider using a DRL-based scheme integrated with the FL paradigm to solve this problem considering: 1) Users have dynamically changing requests and positions, conventional heuristic schemes are hard to cope with new user requests, in contrast, DRL has strong generalization capabilities, well-trained DRL agents can efficiently deal with the dynamic environment; 2) In a 6G system with massive BSs that continue to grow, the dimensions of the problem will grow with the number of slices, heuristic schemes in this case will spend more time searching for the solutions, which means that it suffers a significant performance degradation in a finite time scale.…”

Energy-Efficient Scheduling in Job Shop Manufacturing Systems: A Literature Review

“…Nevertheless, there are a few exceptions. Some authors [37,38,85,[87][88][89][90][91][92] schedule transport and take into account the energy required to load/unload vehicles, and others [93][94][95][96][97][98] consider the energy required to transport the jobs between geographically distributed facilities.…”

Digital twin oriented multi-objective flexible job shop scheduling model and its hybrid particle swarm optimization

“…Each workpiece J i has a multi-process Q ij , j ∈ { 1 , 2 ,⋯ , n i } , which n i is the total number of processes for the workpiece J i . 25 The process of each workpiece is predetermined, one process can be processed on several equipment, and the processing time and cost are different due to equipment differences. The ultimate goal is to determine the processing sequence of each process, and at the same time, to select the appropriate processing equipment and the processing start time for each process, so as to optimize the comprehensive evaluation results of multiple indicators such as the maximum makespan, total machine load, production cost and carbon emissions.…”