Hybrid genetic algorithm for test bed scheduling problems

In this paper, we address the scheduling problem for a heavy industry company which provides ship engines for shipbuilding companies. Before being delivered to customers, ship engines are assembled, tested and disassembled on the test beds. Because of limited test bed facilities, it is impossible for the ship engine company to satisfy all customers’ orders. Therefore, they must select the orders that can be feasibly scheduled to maximise profit. An integer programming model is developed for order selection and test bed scheduling but it cannot handle large problems in a reasonable amount of time. Consequently, a hybrid genetic algorithm (GA) is suggested to solve the developed model. Several experiments have been carried out to demonstrate the performance of the proposed hybrid GA in scheduling test beds. The results show that the hybrid GA performs with an outstanding run-time and small errors in comparison with the integer programming model.     

Multi-objective genetic algorithm for energy-efficient job shop scheduling

The paper investigates the effects of production scheduling policies aimed towards improving productive and environmental performances in a job shop system. A green genetic algorithm allows the assessment of multi-objective problems related to sustainability. Two main considerations have emerged from the application of the algorithm. First, the algorithm is able to achieve a semi-optimal makespan similar to that obtained by the best of other methods but with a significantly lower total energy consumption. Second, the study demonstrated that the worthless energy consumption can be reduced significantly by employing complex energy-efficient machine behaviour policies.     

Hybrid genetic algorithm for bi-objective resource-constrained project scheduling

In this study, we considered a bi-objective, multi-project, multi-mode resource-constrained project scheduling problem. We adopted three objective pairs as combinations of the net present value (NPV) as a financial performance measure with one of the time-based performance measures, namely, makespan (C_max), mean completion time (MCT), and mean flow time (MFT) (i.e., minC_max/maxNPV, minMCT/maxNPV, and minMFT/maxNPV). We developed a hybrid non-dominated sorting genetic algorithm II (hybrid-NSGA-II) as a solution method by introducing a backward–forward pass (BFP) procedure and an injection procedure into NSGA-II. The BFP was proposed for new population generation and post-processing. Then, an injection procedure was introduced to increase diversity. The BFP and injection procedures led to improved objective functional values. The injection procedure generated a significantly high number of non-dominated solutions, thereby resulting in great diversity. An extensive computational study was performed. Results showed that hybrid-NSGA-II surpassed NSGA-II in terms of the performance metrics hypervolume, maximum spread, and the number of non-dominated solutions. Solutions were obtained for the objective pairs using hybrid-NSGA-II and three different test problem sets with specific properties. Further analysis was performed by employing cash balance, which was another financial performance measure of practical importance. Several managerial insights and extensions for further research were presented.     

A genetic algorithm approach for production scheduling with mould maintenance consideration

The traditional approach for maintenance scheduling concerns single-resource (machine) maintenance during production which may not be sufficient to improve production system reliability as a whole. Besides, in the literature many researchers schedule maintenance activities periodically with fixed maintenance duration. However, in a real manufacturing system maintenance activities can be executed earlier and the maintenance duration will become shorter since less time and effort are required. A practical example is that in a plastic production system, the proportion of machine-related downtime is even lower than mould-related downtime. The planned production operations are usually interrupted seriously because of the mismatch among the maintenance periods between injection machine and mould. In this connection, this paper proposes to jointly schedule production and maintenance tasks of multi-resources in order to improve production system reliability by reducing the mismatch among various processes. To integrate machine and mould maintenance tasks in production, this paper attempts to model the production scheduling with mould scheduling (PS-MS) problem with time-dependent deteriorating maintenance schemes. The objective of this paper is to propose a genetic algorithm approach to schedule maintenance tasks jointly with production jobs for the PS-MS problem, so as to minimise the makespan of production jobs.     

Co-evolutionary algorithm: An efficient approach for bilevel programming problems

The bilevel programming problem involves two optimization problems, which is hierarchical, strongly NP-hard and very challenging for most existing optimization approaches. An efficient universal co-evolutionary algorithm is developed in this article to deal with various bilevel programming problems. In the proposed algorithm, evolutionary algorithms are used to explore the leader's and the follower's decision-making spaces interactively. Unlike other existing approaches, in the suggested procedure the follower's problem is solved in two phases. First, an evolutionary algorithm is run for a few generations to obtain an approximation of lower level solutions. In the second phase, from all approximate solutions obtained above, only a small number of good points are selected and evolved again by a newly designed multi-criteria evolutionary algorithm. The technique refines some candidate solutions and can efficiently reduce the computational cost of obtaining feasible solutions. Proof-of-principle experiments demonstrate the efficiency of the proposed approach.     

An adaptive multi-population genetic algorithm for job-shop scheduling problem

Job-shop scheduling problem (JSP) is a typical NP-hard combinatorial optimization problem and has a broad background for engineering application. Nowadays, the effective approach for JSP is a hot topic in related research area of manufacturing system. However, some JSPs, even for moderate size instances, are very difficult to find an optimal solution within a reasonable time because of the process constraints and the complex large solution space. In this paper, an adaptive multi-population genetic algorithm (AMGA) has been proposed to solve this problem. Firstly, using multi-populations and adaptive crossover probability can enlarge search scope and improve search performance. Secondly, using adaptive mutation probability and elite replacing mechanism can accelerate convergence speed. The approach is tested for some classical benchmark JSPs taken from the literature and compared with some other approaches. The computational results show that the proposed AMGA can produce optimal or near-optimal values on almost all tested benchmark instances. Therefore, we can believe that AMGA can be considered as an effective method for solving JSP.     

A genetic algorithm for flexible job shop scheduling with fuzzy processing time

This paper presents a flexible job shop scheduling problem with fuzzy processing time. An efficient decomposition-integration genetic algorithm (DIGA) is developed for the problem to minimise the maximum fuzzy completion time. DIGA uses a two-string representation, an effective decoding method and a main population. In each generation, DIGA decomposes the chromosomes of the main population into a job sequencing part and a machine assigning part and independently evolves the populations of these parts. Some instances are designed and DIGA is tested and compared with other algorithms. Computational results show the effectiveness of DIGA.     

Production scheduling across multi-machine packaging lines for the pre-scheduling level

Scheduling packaging lines in the process industry is a well known hard problem. The problem becomes even more complicated when each line consists of several machines and an order has to be manufactured on different lines. In this paper we consider the case where a number of orders has to switch from one line to another. This means that an order-run is sometimes manufactured first on the first part of a line and then has to switch to the second part of another packaging line. In this situation the second part of the first line and the first part of the second line cannot be used simultaneously for another run. This situation is studied for the case of the tobacco company Royal Kabat BV. In order to solve the complicated planning problem, the process is partitioned into a Pre-scheduling Level (PSL) and a Scheduling Level (SL). At the PSL an aggregated plan for a number of successive weeks is made that takes care of the purchasing of packaging materials and the balancing of the capacities. The PSL plan for the current week is the input for the SL at which the actual packaging schedule for that week is made, including the allocation of orders to lines. In this paper we concentrate on the PSL, and our main purpose is to describe the complicated model building process. The model is solved by means of an improvement heuristic on the solution of an LP-relaxation of the model. The computational results show that our partitioning of the problem into a PSL and an SL yields a practically solvable mathematical model; its solutions (i.e. PL plans) are usable for the SL planning and result in practically useful schedules.     

Modelling and performance of distributed algorithm for scheduling dissimilar machines with set-up

Distributed arrival time control is a highly decentralized scheduling approach where each part entity autonomously controls its arrival time to meet the due-date in real time. This paper presents differential equation-based models for distributed arrival time control of parallel dissimilar machines including sequence-dependent set-up and flowshop scheduling. The main objective was to show that the behaviour of general systems under distributed arrival time control was predictable. Convergence properties of the resulting nonlinear systems were established using the theory of discontinuous differential equations. Geometry was used to gain insight into the behaviour of these nonlinear systems. An approximation model was proposed for mean arrival times when the dynamics resulted in a non-unique steady-state. The model was tested using numerical simulation and agreed well. Geometric insights were also used to investigate scheduling performance of distributed arrival time control. Simulation results indicated that distributed arrival time control could provide significant improvement, typically more than 20%, over commonly used dispatching rules for due-date-based measures. Improved predictability and favourable performance made distributed arrival time control an attractive approach for decentralized control of Just-In-Time production.     

An analogue genetic algorithm for solving job shop scheduling problems

This paper develops a genetic algorithm for solving job shop scheduling problems. It discusses the difficulties arising from the traditional encoding of the problem and suggests a new encoding scheme. The paper also develops an analogue electrical system to represent the problem and uses the measure of that system to develop a new measure for the fitness function of the genetic algorithm. The algorithm considers the conventional genetic operations but with some modification. The computational results, developed for the makespan criterion, show that, for this criterion, the algorithm is reliable and performs relatively well.     

Hybrid genetic algorithm for group technology economic lot scheduling problem

The concept of group technology has been successfully applied to many production systems, including flexible manufacturing systems. In this paper we apply group technology principles to the economic lot scheduling problem, which has been studied for over 40 years. We develop a heuristic algorithm and a hybrid genetic algorithm for the group technology economic lot scheduling problem. Numerical experiments show that the developed algorithms outperform the existing heuristics.     

A workload balancing genetic algorithm for the quay crane scheduling problem

This paper proposes a novel genetic algorithm to deal with the quay crane scheduling problem (QCSP), which is known to be one of the most critical tasks in terminal operations because its efficiency and the quality of the schedule directly influence the productivity of the terminal. QCSP has been studied intensively in recent years. Algorithms in this field are concerned in the solution quality obtained and the required computational time. As QCSP is known to be NP-hard, heuristic approaches are widely adopted. The genetic algorithm proposed is constructed with a novel workload balancing heuristics, which is capable of considering the loading conditions of different quay cranes (QCs) during the reassignment of task-to-QC. The idea is modelled as a fuzzy logic controller to guide the mutation rate and mutation mechanism of the genetic algorithm. As a result, the proposed algorithm does not require any predefined mutation rate. Meanwhile, the genetic algorithm can more adequately reassign tasks to QCs according to the QCs’ loading condition throughout the evolution. The proposed algorithm has been tested with the well-known benchmark problem sets in this field and produces some new best solutions in a much shorter computational time.     

Dynamic scheduling of FMS using a real-time genetic algorithm

The paper presents a genetic algorithm capable of generating optimised production plans in flexible manufacturing systems. The ability of the system to generate alternative plans following part-flow changes and unforeseen situations is particularly stressed (dynamic scheduling). Two contrasting objectives represented by the reduction of machine idle-times, thanks to dynamic scheduling computation and the reduction of the makespan, are taken into account by the proposed system. The key-point is the real-time response obtained by an optimised evolutionary strategy capable of minimising the number of genetic operations needed to reach the optimal schedule in complex manufacturing systems.     

Research on job-shop scheduling problem based on genetic algorithm

With job-shop scheduling (JSS) it is usually difficult to achieve the optimal solution with classical methods due to a high computational complexity (NP-hard). According to the nature of JSS, an improved definition of the JSS problem is presented and a JSS model based on a novel algorithm is established through the analysis of working procedure, working data, precedence constraints, processing performance index, JSS algorithm and so on. A decode select string (DSS) decoding genetic algorithm based on operation coding modes, which includes assembly problems, is proposed. The designed DSS decoding genetic algorithm (GA) can avoid the appearance of infeasible solutions through comparing current genes with DSS in the decoding procedure to obtain working procedure which can be decoded. Finally, the effectiveness and superiority of the proposed method is clarified compared to the classical JSS methods through the simulation experiments and the benchmark problem.     

A genetic algorithm based procedure for more realistic job shop scheduling problems

This work presents a robust procedure to solve job shop scheduling problems with large number of more realistic constraints such as jobs with several subassembly levels, alternative processing plans for parts and alternative resources for operations, requirement of multiple resources to process an operation (e.g. machine, tools, fixtures, staff), resource calendars, batch overlap and sequence dependent setups. Also, the approach considers multi-objective evaluation functions. The system uses modified schedule generation algorithms to obtain a set of initial solutions. Each initial solution is enhanced by a local improvement procedure. Then a hybrid genetic algorithm, which incorporates a local hill climbing procedure, is applied to the set of local optimum schedules.     

Minimisation of total tardiness for identical parallel machine scheduling using genetic algorithm

In recent years research on parallel machine scheduling has received an increased attention. This paper considers minimisation of total tardiness for scheduling of n jobs on a set of m parallel machines. A spread-sheet-based genetic algorithm (GA) approach is proposed for the problem. The proposed approach is a domain-independent general purpose approach, which has been effectively used to solve this class of problem. The performance of GA is compared with branch and bound and particle swarm optimisation approaches. Two set of problems having 20 and 25 jobs with number of parallel machines equal to 2, 4, 6, 8 and 10 are solved with the proposed approach. Each combination of number of jobs and machines consists of 125 benchmark problems; thus a total for 2250 problems are solved. The results obtained by the proposed approach are comparable with two earlier approaches. It is also demonstrated that a simple GA can be used to produce results that are comparable with problem-specific approach. The proposed approach can also be used to optimise any objective function without changing the basic GA routine.     

Power distribution network expansion scheduling using dynamic programming genetic algorithm

A genetic algorithm that is dedicated to the expansion planning of electric distribution systems is presented, with incremental expansion scheduling along a time horizon of several years and treated as a dynamic programming problem. Such a genetic algorithm (called dynamic programming genetic algorithm) is endowed with problem-specific crossover and mutation operators, dealing with the problem through a heuristic search in the space of dynamic programming variables. Numerical tests have shown that the proposed algorithm has found good solutions that considerably enhance the solutions found by non-dynamic programming methods. The algorithm has also shown to work for problem sizes that would be computationally infeasible for exact dynamic programming techniques.     

A genetic algorithm for solving the economic lot scheduling problem in flow shops

In this study, we propose a hybrid genetic algorithm (HGA) to solve the economic lot scheduling problem in flow shops. The proposed HGA utilizes a so-called Proc PLM heuristic that tests feasibility for the candidate solutions obtained in the evolutionary process of genetic algorithm. When a candidate solution is infeasible, we propose to use a binary search heuristic to ‘fix’ the candidate solution so as to obtain a feasible solution with the minimal objective value. To evaluate the performance of the proposed HGA, we randomly generate a total of 2100 instances from seven levels of utilization rate ranged from 0.45 to 0.80. We solve each of those 2100 instances by the proposed HGA and the other solution approaches in the literature. Our experiments show that the proposed HGA outperforms traditional methods for solving the economic lot scheduling problem in flow shops.     

遗传算法的双目标柔性作业车间调度研究

研究了FMS环境下先进制造车间路径柔性的优化调度问题.同时考虑现代生产准时制的要求,建立了柔性作业车间调度问题的双目标数学优化模型,并给出了求解模型的遗传算法的具体实现过程;针对模型的特殊性,提出了染色体两层编码结构,将AOV网络图应用到解码和适应度函数的计算中,通过一个调度实例进行验证,给出了相应的选择、交叉、变异操作设计方案.     

A genetic algorithm approach to scheduling PCBs on a single machine

This paper addresses the problem of scheduling N printed circuit boards (PCBs) on a single machine equipped with an automatic component interchange mechanism. Assume that the total number of different components required to process all N PCBs is greater than the capacity of the spool. If the requisite components are not on the spool, then one or more component switches must occur before the PCB can be processed. The problem consists of finding the order to schedule the PCBs on the axial insertion machine and the components to place on the spool before each PCB is processed. The performance criterion is to minimize the total number of component switches. This problem is addressed employing a genetic algorithm to search the space of alternative solutions. To evaluate the performance of the GA, a heuristic solution based on a travelling salesman formulation is described. Extensive experiments were carried out for both approaches based on data extracted from industrial scenes.