Research on flexible job-shop scheduling problem based on a modified genetic algorithm

Abstract

Aiming at the existing problems with GA (genetic algorithm) for solving a flexible job-shop scheduling problem (FJSP), such as description model disunity, complicated coding and decoding methods, a FJSP solution method based on GA is proposed in this paper, and job-shop scheduling problem (JSP) with partial flexibility and JIT (just-in-time) request is transformed into a general FJSP. Moreover, a unified mathematical model is given. Through the improvement of coding rules, decoding algorithm, crossover and mutation operators, the modified GA's convergence and search efficiency have been enhanced. The example analysis proves the proposed methods can make FJSP converge to the optimal solution steadily, exactly, and efficiently.

A Taxonomy for the Flexible Job Shop Scheduling Problem

“…Zhang et al [38,39], Gao et al [40][41][42], Li et al [43,44], Wang et al [45], and Xing et al [46] used similar representations as that of Ho et al [27], except that the machine selection component was constructed with integer values instead of binary values. Sun et al [47] used integer values for the operation sequence component that showed the position on the schedule. Frutos et al [48] used integer values for both components, in which a gene on the operation sequence component represented a possible order of operations on each machines.…”

Solving the flexible job-shop just-in-time scheduling problem with quadratic earliness and tardiness costs

“…In the studies reviewed, the FJSSP has been addressed using only machines [31,33], jobs [27], and operations as decision variables [12,23]. These variables do not implicitly offer solutions that comply with JIT principles of reducing waste.…”

“…Due to this complexity, the FJSSP has been mainly addressed for makespan minimization or a linear combination of the makespan with other objectives such as machine workload, flow time, and maximum or total tardiness. Generally speaking, when the FJSSP with JIT production objectives has been addressed, a linear relationship between earliness and tardiness costs was used [11,12]. Nevertheless, simply minimizing tardiness costs, or the linear combination of earliness and tardiness, is too simplistic since tardiness costs comprise other qualitative indicators related to customer loyalty (i.e., customer dissatisfaction and the risk of losing the customer).…”

Reducing myopic behavior in FMS control: A semi-heterarchical simulation–optimization approach

“…Generally speaking, when JIT production objectives are considered, the scheduling models in the literature assume a linear relationship between earliness and tardiness costs (Guo et al, 2008;Sun et al, 2010;Gomes et al, 2005;El Khoukhi et al, 2011). Nevertheless, simply minimizing tardiness costs, or the linear combination of earliness and tardiness, is too simplistic since tardiness costs comprise other qualitative indicators related to customer loyalty (i.e., customer dissatisfaction and the risk of losing the customer).…”