Machine Scheduling Problem Research Articles

Novel mathematical formulations for parallel-batching processing machine scheduling problems

We study mathematical formulations for batch-processing machine scheduling problems (BPMPs), which are the challenging issues in the machine scheduling literature where machines are capable of processing a batch of jobs simultaneously if jobs with non-identical sizes can be packed in a capacitated machine. In this paper, we tackle single- and parallel-machine BPMPs, and other interesting problem variants that aim at minimizing the makespan. We develop novel formulations along with valid inequalities and an algorithm framework that makes use of dual information and bounding techniques to achieve efficiency when instances are intractable. Extensive computational experiments on benchmark instances show that our approaches achieve state-of-the-art results and prove the optimality of intractable instances in the literature.

Lower and upper bounds for scheduling a real-life assembly problem with precedences and resource constraints

In this work, we consider a real-life scheduling problem involving the production of off-road vehicles using a three-level assembly process, subject to precedence, machines, and resource constraints. This problem shares multiple characteristics with other scheduling problems such as the Parallel Machine Scheduling and Flexible Flow Shop Problems. However, it also comprises less investigated aspects such as a specific job precedence structure resulting in a directed rooted in-tree and a global resource constraint limiting the number of simultaneously active machines. We present a straightforward adaptation of a time-indexed mathematical formulation to the problem and introduce a new lower-bounding procedure. To solve the problem, we propose constructive heuristics based on classical priority rules and adaptations of job sequencing heuristics. We also propose CORE, a specialized approach tailored to leverage the problem structure. All the algorithms are extensively evaluated on two benchmark sets, various shop floor configurations, and eight real-life scenarios. Our results reveal the general effectiveness of job sequencing approaches and demonstrate the overall superiority of CORE.

A new composite heuristic to minimize the total tardiness for the single machine scheduling problem with variable and flexible maintenance

Inspired by a real-world maintenance/job scheduling issue coming from the semiconductor industry, the present paper proposes a new heuristic algorithm structure for the single machine scheduling T-problem with flexible and variable maintenance, job release dates and sequence dependence setup times. Considering the typical short-term production planning needs, a single maintenance problem has to be scheduled within a certain time interval, along with a set of jobs so as to minimize the total tardiness. A twofold contribution emerges from the present paper. First, four mixed-integer linear programming models are developed for the problem at hand and compared in terms of time to convergence and computational complexity. Second, a novel heuristic algorithm, which has been configured into three distinct variants, has been compared with 17 alternative heuristics from the relevant literature based on a comprehensive experimental campaign. The numerical results allow the selection of the most suitable MILP model and confirm the effectiveness of the proposed heuristic approach.

Minimizing total completion time and makespan for a multi-scenario bi-criteria parallel machine scheduling problem

Multi-criteria scheduling problems under uncertainty remain a relatively unexplored topic in theoretical computer science despite substantial practical interests. This work studies a bi-objective identical parallel machine scheduling problem under uncertainty, in which the first objective is to minimize the total completion time, and the second is to minimize the makespan. Especially a job’s processing time is assumed to be represented by a polynomial function with respect to scenario u∈U, where U⊂R+ is an interval containing an infinite number of scenarios.In this work, we are looking for a compact and complete description of the set of possibly optimal solutions, along with their objective function values, over the set of scenarios or an approximation with a performance guarantee. First, to better understand the characteristics of the studied problem, we consider two single-objective problems: parallel machine scheduling problem under uncertainty with the total completion time and makespan criterion, respectively. For the problem with the total completion time criterion, we demonstrate that the set of possibly optimal schedules can be found in polynomial time. In contrast, for the problem with the makespan criterion, we provide a (1+ϵ)-approximation algorithm. For the bi-objective problem, we provide a 2-approximation algorithm for any number of parallel machines and a (1+ϵ)-approximation algorithm where a fixed number of parallel machines is considered.

Single machine scheduling with the total weighted late work and rejection cost

AbstractWe study a single machine scheduling problem with the total weighted late work and the total rejection cost. The late work of a job is the part of this job executed after its due date, and the rejection cost of a job is the fee of rejecting to process it. We consider a Pareto scheduling and two restricted scheduling problems. The Pareto scheduling problem aims to find all non‐dominated values of the total weighted late work and the total rejection cost. The restricted scheduling problems are dedicated to minimizing the total weighted late work with the total rejection cost not greater than a given threshold, and minimizing the total rejection cost with the total weighted late work not greater than a given threshold, respectively. For the Pareto scheduling problem, we prove that it is binary NP‐hard by providing a pseudo‐polynomial time algorithm, and give a fully polynomial time approximation scheme (FPTAS). For the restricted scheduling problems, we prove that there are no FPTASes unless , which answers an open problem. Moreover, we develop relaxed FPTASes for these two restricted scheduling problems.

A transformer-based deep reinforcement learning approach for dynamic parallel machine scheduling problem with family setups

A transformer-based deep reinforcement learning approach for dynamic parallel machine scheduling problem with family setups

The Parallel Machine Scheduling Problem with Different Speeds and Release Times in the Ore Hauling Operation

The Parallel Machine Scheduling Problem with Different Speeds and Release Times in the Ore Hauling Operation

Parallel machine scheduling with a total energy consumption limitation for minimizing total completion time

This article investigates an energy-efficient, identical parallel machine scheduling problem. The objective is to minimize the total completion time while total energy costs do not exceed a given limitation. First, a reasonable range of total energy cost limits and the NP-hardness of this problem are analysed. Then, a mixed-integer programming model is presented. Afterwards, an improved simulated annealing (ISA) algorithm is devised. The ISA algorithm incorporates four effective neighbourhood operators, a method for adaptive selection of operators, and an initial feasible solution generation method. Comparison with genetic algorithms (GAs) and simplified swarm optimization (SSO) algorithms verifies the effectiveness and efficiency of the ISA algorithm. Results demonstrate that the ISA algorithm can provide near-optimal or better solutions than CPLEX ® in small- and medium-scale instances. Experimental results on large-scale instances show that ISA significantly outperforms GAs and SSOs. ISA, in particular, can find the best values for all instances of 500 jobs except one.

A distributionally robust approach for the parallel machine scheduling problem with optional machines and job tardiness

This paper investigates a parallel machine scheduling problem with uncertain job processing time, where the job tardiness and optional machines are considered. To address the factor of energy saving, only a subset of all available machines are turned on, which is referred to as not-all-machine (NAM). To depict the uncertain processing time, a mean–mean absolute deviation (MAD) ambiguity set is utilized, and the cost of job tardiness is minimized under the worst-case distribution scenario over the ambiguity set. After building a distributionally robust optimization (DRO) model, theoretical bounds of the optimal number of machines are obtained. Since the model is not computationally scalable, an upper bound on its inner minimization problem is employed, and a mixed integer linear programming (MILP) approximation is obtained based on McCormick inequalities. For the DRO model, tailored speedup techniques are employed, significantly enhancing the computational performance. To evaluate the validity of the proposed DRO model, we compare it with its stochastic programming (SP) counterpart under various parameter settings. Numerical experiments demonstrate that the DRO model exhibits strong performance in the worst-case scenarios. As the problem size increases, the DRO model casts clear advantages over the SP model in terms of computational efficiency and reliability. It is observed that the performance of the DRO model is more stable than that of the nominal sequence, especially with loose due dates. Furthermore, the out-of-sample performance under various decision making preferences shed new lights into the trade-off between energy saving and production efficiency.

Advanced constraint programming formulations for additive manufacturing machine scheduling problems

In comparison with traditional subtractive manufacturing techniques, additive manufacturing (AM) enables fabricating complex parts through a layer-by-layer process. AM makes it possible to produce one-piece and lightweight functional products, which are traditionally made from several parts. This paper introduces constraint programming (CP) models to minimise makespan in single, parallel identical and parallel unrelated AM machine scheduling environments for selective laser melting. Alternative CP formulations are explored to improve efficiency. The proposed CP model significantly benefits from the introduction of interval variables to replace binary assignment variables, and pre-definitions to narrow the search space, resulting in increased search performance. A computational study has been conducted to compare the performance of our proposed CP model with both a mixed-integer programming and a genetic algorithm from existing literature, evaluating improvements made to its search capability. Computational results indicate that the proposed CP model can obtain high-quality solutions in a timely manner even for several large-size instances.

An incremental learning approach to dynamic parallel machine scheduling with sequence-dependent setups and machine eligibility restrictions

Minimizing the tardiness of a parallel machine scheduling problem has been actively studied in modern manufacturing systems. In particular, dynamic parallel machine scheduling problems (DPMSPs) have gained much attention since rescheduling is required to address unpredictable events such as unexpected job arrivals and machine breakdowns, which usually occur in real-world manufacturing systems. To deal with such unpredictable events, many researchers have employed deep supervised learning (SL). However, it is still challenging to solve a DPMSP since SL requires a large number of high-quality schedules for the training neural networks. In this paper, we propose an incremental learning-based scheduling method (ILS) in which neural networks (NNs) are periodically trained by utilizing schedules built from the updated NNs in previous training intervals. Furthermore, to perform training without reducing solution space within a dynamic environment, the proposed method is designed to consider the dependency between consecutive allocations on a machine for the sum of setup time and tardiness. To verify the effectiveness of the proposed method, extensive experiments are carried out in static and dynamic scheduling problems. The experiment results demonstrate that ILS outperforms the existing methods such as dispatching rules, metaheuristics, and DRL-based methods. Additionally, we show that the proposed input features are effective in appropriately selecting job-machine pairs for assigning jobs on eligible machines through deep Shapley additive explanations analysis.

Just-in-time scheduling of unrelated parallel machines with family setups and soft time window constraints

ABSTRACT This study addresses an unrelated parallel machine scheduling problem with family setups and soft time windows, which includes considerations for machine eligibility. The objective is to minimize the total weighted tardiness and earliness, and to maximize the number of Just-in-Time jobs. Four different metaheuristic algorithms are compared to determine which type – single-point (simulated annealing), population-based evolutionary (artificial immune system, genetic algorithm), or swarm intelligence algorithms (ant colony optimization) – is more effective in solving this problem. Two strategies, namely penalty and repair, are proposed to handle precedence constraints. Experimental results demonstrate that ant colony optimization with local search which employs the repair strategy, as a swarm intelligence-based algorithm outperforms its rivals in terms of objective function values and robustness. The performance of this algorithm has shown an improvement in objective function values ranging from 2.12% to 8.52% on average, with a relative percentage deviation of 10.35.

A restless bandit approach for capacitated condition based maintenance scheduling

AbstractThis paper considers the maintenance scheduling problem of multiple non-identical machines deteriorating over time. The deterioration gradually decreases a machine’s performance, which results in revenue losses due to lower output quality. The maintenance cost is dependent on the degradation state, and the number of maintenance activities that can be carried out simultaneously is restricted by the number of maintenance workers. Our main goal is to propose a heuristic with low complexity that consistently produces solutions close to the optimal strategy for problems of real size. We cast the problem as a restless bandit problem and propose an index based heuristic (Whittle’s index policy) which can be computed efficiently. We also provide a lower bound that can be computed by linear programming. We numerically compare the performance of the index heuristic with alternative policies. In addition to achieving superior performance over failure-based and threshold policies, Whittle’s policy numerically converges to our lower bound when the number of machines is moderately high and/or maintenance workload is high.

Solving tri-criteria: total completion time, total late work, and maximum earliness by using exact, and heuristic methods on single machine scheduling problem

The presented study investigated the scheduling regarding jobs on a single machine. Each job will be processed with no interruptions and becomes available for the processing at time 0. The aim is finding a processing order with regard to jobs, minimizing total completion time , total late work , and maximal tardiness which is an NP-hard problem. In the theoretical part of the present work, the mathematical formula for the examined problem will be presented, and a sub-problem of the original problem of minimizing the multi-objective functions is introduced. Also, then the importance regarding the dominance rule (DR) that could be applied to the problem to improve good solutions will be shown. While in the practical part, two exact methods are important; a Branch and Bound algorithm (BAB) and a complete enumeration (CEM) method are applied to solve the three proposed MSP criteria by finding a set of efficient solutions. The experimental results showed that CEM can solve problems for up to jobs. Two approaches of the BAB method were applied: the first approach was BAB without dominance rule (DR), and the BAB method used dominance rules to reduce the number of sequences that need to be considered. Also, this method can solve problems for up to , and the second approach BAB with dominance rule (DR), can solve problems for up to jobs in a reasonable time to find efficient solutions to this problem. In addition, to find good approximate solutions, two heuristic methods for solving the problem are proposed, the first heuristic method can solve up to jobs, while the second heuristic method can solve up to jobs. Practical experiments prove the good performance regarding the two suggested approaches for the original problem. While for a sub-problem the experimental results showed that CEM can solve problems for up to jobs, the BAB without dominance rule (DR) can solve problems for up to , and the second approach BAB with dominance rule (DR), can solve problems for up to jobs in a reasonable time to find efficient solutions to this problem. Finally, the heuristic method can solve up to jobs. Arithmetic results are calculated by coding (programming) algorithms using (MATLAB 2019a)

Novel integer L-shaped method for parallel machine scheduling problem under uncertain sequence-dependent setups

We study scheduling problems on unrelated parallel machines with uncertainty in job processing and sequence-dependent setup times. We first formulate this problem as a two-stage stochastic program using a scenario-based approach, in which the first-stage decisions involve job-machine assignments, and the second-stage deals with job sequencing. We then propose an innovative modified integer L-shaped method, uniquely designed to decompose the primary problem into independent subproblems at both the scenario and machine levels. We have augmented this method with a specialized integer optimality cut, which is further strengthened using a sequential lifting procedure. To broaden the applicability of our model and solution, we have adapted it to a distributionally robust optimization (DRO) framework under Wasserstein ambiguity. This process involves reformulating the DRO model into a mixed-integer linear program (MILP) using conic duality theory. Additionally, we integrate a distribution separation program to define the worst-case probability distribution during each iteration of the proposed integer L-shaped method, enhancing the computational speed for solving the DRO model. We conducted a series of numerical experiments to illustrate the effectiveness and scalability of the proposed solution methodology, which demonstrated highly promising results. Our innovative decomposition algorithms showed superior performance, achieving solution times that are significantly faster (up to two orders of magnitude) compared to the traditional integer L-shaped method for the stochastic scheduling model and the direct MILP reformulation of the DRO model. These results were obtained for problem instances solved to the exactness within the time limit. Moreover, the solution derived from the DRO model exhibited greater robustness in out-of-sample analysis with simulated data compared to the solution obtained from the stochastic model.

Variable neighbourhood search for parallel machine scheduling with a single loading server: a truck-scheduling perspective

ABSTRACT In this paper, we study a class of parallel machine scheduling problem with a single loading and multiple unloading servers. We show an application of this problem in a truck-scheduling context. We design a mixed-integer-programming formulation and use it to solve instances of various sizes with different characteristics. The obtained results reveal that a state-of-the-art commercial solver is only able to tackle instances of limited size. To solve larger instances, a general variable neighborhood search is proposed. Our comprehensive computational experiments demonstrate how the proposed metaheuristic can provide high-quality solutions in reasonable computing time (a few seconds). This type of performance is particularly relevant when considering operational problems that must be tackled on a daily basis. Finally, a sensitivity analysis is performed with respect to the number of trucks.

A note on scheduling identical parallel machines with preemptions and setup times

We correct, extend and improve recent algorithmic and computational complexity results for the identical parallel machine scheduling problem with preemptions and setup times preceding uninterrupted job parts. The objective is to minimise the makespan. Our results include a proof of strong NP-hardness for an arbitrary number of machines, and a pseudo-polynomial algorithm and an FPTAS for the two machines case.

Optimization of the production planning process in a nuts and dried fruits industry

A resource-constrained machine scheduling problem is addressed in this paper. The problem arises in an industry located in the Canary Islands (Spain) and is characterized by a set of jobs to be completed on multiple processing lines, where each job has a specific processing time and a sequence-dependent setup time. The objective is to maximize the productivity of the processing lines, subject to constraints such as availability of personnel and processing lines, due dates, precedence order, and setup times, among others.In order to solve this problem, an optimization model that represents the problem as a mixed-integer program is proposed. The paper also introduces a constraint satisfaction heuristic to find near-optimal solutions of the problem. The effectiveness of the proposal is demonstrated through computational experiments on a set of real and synthetic instances, showing that the proposed technique is able to find solutions that are close to the optimal ones within a reasonable amount of time.

Combinatorial optimization methods for yarn dyeing planning

AbstractManaging yarn dyeing processes is one of the most challenging problems in the textile industry due to its computational complexity. This process combines characteristics of multidimensional knapsack, bin packing, and unrelated parallel machine scheduling problems. Multiple customer orders need to be combined as batches and assigned to different shifts of a limited number of machines. However, several practical factors such as physical attributes of customer orders, dyeing machine eligibility conditions like flotte, color type, chemical recipe, and volume capacity of dye make this problem significantly unique. Furthermore, alongside its economic aspects, minimizing the waste of natural resources during the machine changeover and energy are sustainability concerns of the problem. The contradictory nature of these two makes the planning problem multi-objective, which adds another complexity for planners. Hence, in this paper, we first propose a novel mathematical model for this scientifically highly challenging yet very practical problem from the textile industry. Then we propose Adaptive Large Neighbourhood Search (ALNS) algorithms to solve industrial-size instances of the problem. Our computational results show that the proposed algorithm provides near-optimal solutions in very short computational times. This paper provides significant contributions to flexible manufacturing research, including a mixed-integer programming model for a novel industrial problem, providing an effective and efficient adaptive large neighborhood search algorithm for delivering high-quality solutions quickly, and addressing the inefficiencies of manual scheduling in textile companies; reducing a time-consuming planning task from hours to minutes.

Scheduling with jobs at fixed positions

In this paper, we study classical single machine scheduling problems with the additional constraint that a set of special jobs must be scheduled at certain positions in the job sequence. In other words, a special job must start when a certain number of jobs have finished on the machine. We analyze several classical objective functions for this more general setting with the additional constraint of fixed positioned jobs. We show that for some of them they are still polynomially solvable. Then, we focus on the objective of minimizing the number of tardy jobs. Considering the case of just one special job, this allows for a polynomial-time algorithm.