Discrete OptimizationDistributionally robust scheduling on parallel machines under moment uncertainty
Introduction
The parallel machine scheduling problem (PMSP) considers a set of jobs to be processed on several identical machines over a given time period. Each job has to be processed on one of the machines, and no machine can process more than one job at the same time. A feasible schedule contains the allocation of machines and the sequence of jobs on each machine. The aim of PMSP is to find a schedule, which optimizes one or more performance criteria, among all feasible ones. As a decision-making process, it plays an important role in many manufacturing and service industries (Pinedo, 2012). Although PMSP is formulated in the manufacturing environment, this model is suitable for many other fields (Mokotoff, 2004), such as social science (Liao, Van Delft, Vial, 2013, Mak, Rong, Zhang, 2014b) and computer science (Kim, Park, Cui, Kim, & Gruver, 2009). For example in outpatient clinics, the doctors and patients can be regarded as the machines and jobs, respectively. The turnaround time elapsed from a patient’s coming to its leaving, which is the key performance measure of appointment systems, is related to the flow time of a job in PMSP. The deterministic PMSP has been extensively studied in the past decades (Cheng, Sin, 1990, Edis, Oguz, Ozkarahan, 2013, Vallada, Ruiz, 2011). However, the job processing times are always uncertain because of the personalized customization, the unknown operation time for a new patient, or some other uncertain factors. Under this circumstance, the ‘optimal’ schedules obtained by deterministic models may turn out to be poor in practice (Daniels & Kouvelis, 1995), which motivates us to study scheduling problems with uncertainties.
Popular approaches applied to deal with uncertainties in scheduling are stochastic program (SP) and robust optimization (RO), both of which have inherent drawbacks. SP models regard the uncertain parameters as random variables with known distributions (Pinedo, 1983, Skutella, Uetz, 2005, Van Der Heyden, 1981, Weber, Varaiya, Walrand, 1986), but the exact distributions are hard to obtain due to limited information in practice. While RO models (Daniels, Kouvelis, 1995, Tadayon, Smith, 2015, Yang, Yu, 2002) restrict the uncertain parameters to given uncertainty sets without any distributional assumption, which fails to make full use of the information that may be obtained via historical data. To overcome these shortcomings, we propose a distributionally robust (DR) formulation for PMSP, where uncertain parameters are regarded as random variables subject to an ambiguous distribution limited in a distributional set. Besides the support of uncertain parameters, the mean and covariance matrix are also involved in the formulation, which makes the DR models contain more information than the RO ones.
In this paper, we study an identical PMSP with uncertain job processing times in DR framework. The objective functions of PMSP are generally related to the completion times or the due dates of different jobs, such as the maximum completion time, i.e., makespan (Koulamas, Kyparisis, 2009, Mokotoff, 2004), the total flow time (Li, Yang, 2009, Liaw, 2016, Weng, Lu, Ren, 2001), the total/maximum tardiness (Bilge, Kiraç, Kurtulan, Pekgün, 2004, Lee, 2017) and the number of tardy jobs (Ho & Chang, 1995). Because the flow time of each job means the time elapsed from the job’s release to its completion, the total flow time (TFT) indicates the total holding or inventory costs incurred by a schedule. Among different metrics, TFT is widely accepted as a good measure for the overall quality of production. Furthermore, it is linear in job processing times, which is a favorable property for our model transformation. Therefore, we choose TFT as the performance measure and develop the DR formulation based on a distributional set specified by the support and first two moments information of job processing times. To consider the uncertainties in moments, our distributional set restricts the mean vector in an ellipsoid and the covariance matrix in a positive semidefinite cone. The original formulation is a two-stage min-max problem, which aims at finding a robust schedule with minimal expected TFT under the worst-case distribution. By solving the inner maximization subproblem, the min-max formulation is finally reduced to an integer second-order cone program (I-SOCP).
In general, I-SOCP can be solved by some commercial solvers, such as IBM® ILOG® CPLEX and Mosek, via branch-and-bound based methods. However, these methods are time-consuming when the size of problem instances is large. In production environments, scheduling decisions should be made for each work shift with the time-frame usually being days or hours. However, rescheduling is often required to modify schedules due to uncertainties in the production process, and has to be completed within a much shorter computational time limit. To solve our I-SOCP with higher efficiency, we propose an exact algorithm based on its structural features, which can find out all the solutions satisfying the necessary optimality condition (i.e., NOC-points) and reserve the best one as the optimal solution. This algorithm is intuitively named as NOC-point search algorithm (NPSA). The optimality of NPSA is rigorously proved, and its efficiency is demonstrated by solving large problem instances with up to 100 jobs in a few seconds.
The main contributions in this paper are summarized as follows:
- 1.
We propose a DR optimization approach to model the PMSP with uncertain parameters, which avoids the shortcomings of existing SP and RO models.
- 2.
We consider the uncertainty of estimated moments in our DR formulation, which hedges against the bias caused by insufficient historical data in practice.
- 3.
We reduce the original min-max DR formulation to an I-SOCP problem.
- 4.
We design an exact algorithm for our I-SOCP problem, which is efficient and easy to implement. The design ideas can be extended to solve other I-SOCP problems with similar structure in different applications.
The rest of this paper is organized as follows. Related literature is reviewed in Section 2, and the formulation of PMSP as well as its DR counterpart are introduced in Section 3. Next in Section 4, we propose the exact algorithm NPSA and provide its optimality proof. In Section 5, we illustrate the empirical performance of NPSA, and analyze the robustness of our DR-PMSP model. Finally in Section 6, we conclude this paper with a few final remarks.
Section snippets
Related literature
A common approach to deal with uncertainties in production scheduling problems is stochastic programming (SP). These SP models regard the uncertain parameters as random variables with known distributions, and usually aim at optimizing the expected performance. For example, Pinedo (1983) developed four SP models, where the job processing times are assumed to be exponentially distributed. Three of them are single machine scheduling problems, and the other one is a PMSP with the expected number of
Deterministic identical parallel machine scheduling
The identical PMSP considers a set of J jobs to be processed on a set of M identical machines. A fundamental assumption is that each machine can process one job at a time, and a job can be processed by any but only one machine non-preemptively. The processing time of job j on each machine is and all the jobs are released at the beginning of the scheduling period. The aim of this problem is to find the best schedule S, which indicates the assignment of these J
Solution algorithm
The main idea of our solution algorithm is to find all the NOC-points of CRP and reserve the best one as the optimal solution. We first design an alternate descent procedure (ADP) to find out NOC-points in Section 4.1. Based on ADP, we next propose the exact algorithm, NOC-point search algorithm (NPSA), in Section 4.2. Finally in Section 4.3, the optimality of NPSA is rigorously proved.
Experimental analyses and insights
In this section, we study the experimental performance of NPSA and the robustness of the proposed DR-PMSP model. Firstly, NPSA is compared with its simplified version Bi-ADP and the commercial solver IBM® ILOG® CPLEX (version V12.6.2 for Windows x64) in Section 5.1. Then, Section 5.2 explores how the number of NOC-points and the relationship between μs and σs influence the performance of NPSA. And finally in Section 5.3, the robustness analyses are given for DR-PMSP3 by comparing its robust
Conclusions
In this paper, we discuss the uncertain parallel machine scheduling problem (PMSP) in the framework of distributionally robust optimization. Based on the deterministic formulation of PMSP, we propose its distributionally robust counterpart, DR-PMSP, to optimize the average performance of TFT under the worst-case distribution. In DR-PMSP, uncertain job processing times are regarded as a random vector subject to an ambiguous distribution. The distributional set relies on the moment information
Acknowledgments
The authors thank Qie He for valuable discussions. This work is supported by the National Natural Science Foundation of China under Grants U1660202, 61473141 and 61503211.
References (61)
- et al.
Complexity of single machine scheduling problems under scenario-based uncertainty
Operations Research Letters
(2008) Parallel machine scheduling with fuzzy processing times using a robust genetic algorithm and simulation
Information Sciences
(2011)- et al.
A tabu search algorithm for parallel machine total tardiness problem
Computers & Operations Research
(2004) - et al.
Distributionally robust single machine scheduling with risk aversion
European Journal of Operational Research
(2017) - et al.
A state-of-the-art review of parallel-machine scheduling research
European Journal of Operational Research
(1990) - et al.
Complexity of interval minmax regret scheduling on parallel identical machines with total completion time criterion
Operations Research Letters
(2016) - et al.
Parallel machine scheduling with additional resources: Notation, classification, models and solution methods
European Journal of Operational Research
(2013) - et al.
A family of inequalities valid for the robust single machine scheduling polyhedron
Computers & Operations Research
(2010) - et al.
Recent advances in robust optimization: An overview
European Journal of Operational Research
(2014) - et al.
Minimizing the number of tardy jobs for m parallel machines
European Journal of Operational Research
(1995)
A modified LPT algorithm for the two uniform parallel machine makespan minimization problem
European Journal of Operational Research
Non-identical parallel-machine scheduling research with minimizing total weighted completion times: Models, relaxations and algorithms
Applied Mathematical Modelling
Minimizing worst-case regret of makespan on a single machine with uncertain processing and setup times
Applied Soft Computing
Optimisation of fault-tolerant fabric-cutting schedules using genetic algorithms and fuzzy set theory
European Journal of Operational Research
An exact algorithm for the identical parallel machine scheduling problem
European Journal of Operational Research
Optimal fuzzy counterparts of scheduling rules
European Journal of Operational Research
Parallel machine scheduling models with fuzzy processing times
Information Sciences
A fuzzy logic based production scheduling/rescheduling in the presence of uncertain disruptions
Fuzzy sets & systems
Two branch-and-bound algorithms for the robust parallel machine scheduling problem
Computers & Operations Research
A genetic algorithm for the unrelated parallel machine scheduling problem with sequence dependent setup times
European Journal of Operational Research
Unrelated parallel machine scheduling with setup consideration and a total weighted completion time objective
International Journal of Production Economics
Stochastic scheduling to minimize expected maximum lateness
European Journal of Operational Research
Managing cost uncertainties in transportation and assignment problems
Advances in Decision Sciences
Robust optimization
Optimal inequalities in probability theory: A convex optimization approach
SIAM Journal on Optimization
A MIP formulation for the minmax regret total completion time in scheduling with unrelated parallel machines
Optimization Letters
β-robust scheduling for single-machine systems with uncertain processing times
IIE transactions
Robust scheduling to hedge against processing time uncertainty in single-stage production
Management Science
Distributionally robust optimization under moment uncertainty with application to data-driven problems
Operations research
Optimization under moment, robust, and data-driven models of uncertainty
Cited by (37)
Target-oriented robust satisficing models for the single machine scheduling problems with release time
2024, Computers and Operations ResearchDistributionally robust scheduling of stochastic knapsack arrivals
2024, Computers and Operations ResearchRobustifying the resource-constrained project scheduling against uncertain durations
2024, Expert Systems with ApplicationsWasserstein distance‐based distributionally robust parallel‐machine scheduling
2023, Omega (United Kingdom)Bi-objective scenario-guided swarm intelligent algorithms based on reinforcement learning for robust unrelated parallel machines scheduling with setup times
2023, Swarm and Evolutionary ComputationSingle machine scheduling with release dates: A distributionally robust approach
2023, European Journal of Operational ResearchCitation Excerpt :The DRO models optimizes the performance measures based on the worst-case distribution within the ambiguity set (Alimoradi et al., 2016; Natarajan, Sim, & Uichanco, 2018; Wang, Zhang, & Tang, 2019). Chang, Ding, & Song (2019) present a min-max DRO model to minimize the expected TFT of the parallel machine scheduling problem, in which the distribution of the uncertain processing time is defined with the moment uncertainty set, and then it is transformed into an integer second-order cone programming (I-SOCP) problem. The DRO model for the SMSP with uncertain processing time and deterministic due date is introduced by Niu et al. (2019), which aims to minimize the expected worst-case total tardiness criterion.