Distributed Production Scheduling Research Articles

A learning-driven multi-objective cooperative artificial bee colony algorithm for distributed flexible job shop scheduling problems with preventive maintenance and transportation operations

In recent years, distributed production scheduling problems receive much attention from both scholars and practicers. Nevertheless, existing research on such problems commonly ignores machine maintenance and job transportation operations. This work proposes a distributed flexible job shop scheduling problem with preventative maintenance and transportation operations in consideration of sequence-dependent setup time. First, a mixed integer programming model is formulated to minimize maximum completion time of jobs and maximum workload of factories. Second, a learning-driven multi-objective cooperative artificial bee colony algorithm is developed to deal with the model. A Q-learning-based cooperation strategy between population and obtained non-dominated individuals is devised to strengthen the exploration and exploitation performance. Furthermore, heuristic-based population initialization, crossover operations in employed bee phase and iterated local search methods in onlooker bee phase are specially developed considering the problem characteristics. Finally, the proposed method is compared against three well-known meta-heuristics from existing literature and an exact solver CPLEX by using a set of benchmark instances. The results confirm the competitive performance of the developed model and algorithm for addressing the studied problems.

A decentralized production–distribution scheduling problem: Solution and analysis

In modern production–distribution supply chains, decentralization has increased significantly, due to increasing production network efficiency. This study investigates a production scheduling and vehicle routing problem in a make-to-order context under a decentralized decision-making structure. Specifically, two different decision makers hierarchically decide the production and distribution schedules to minimize their incurred costs and we formulate the problem as a bi-level mixed-integer optimization model as a static Stackelberg game between manufacturer and distributor. At the upper level, the manufacturer decides its best scheduling under a flexible job-shop manufacturing system, and at the lower level, the distributor decides its distribution scheduling (routing) which influences the upper-level decisions. The model derives the best production–distribution scheduling scheme, with the objective of minimizing the cost of the manufacturer (leader) at the lowest possible cost for the distributor (follower). As the lower level represents a mixed-integer programming problem, it is challenging to solve the resulting bi-level model. Therefore, we extend an efficient decomposition algorithm based on Duplication Method and Column Generation. Finally, to discuss the decentralization value, the results of the presented bi-level model are compared with those of the centralized approach.

N-list-enhanced heuristic for distributed three-stage assembly permutation flow shop scheduling

AbstractSystem-wide optimization of distributed manufacturing operations enables process improvement beyond the standalone and individual optimality norms. This study addresses the production planning of a distributed manufacturing system consisting of three stages: production of parts (subcomponents), assembly of components in Original Equipment Manufacturer (OEM) factories, and final assembly of products at the product manufacturer’s factory. Distributed Three Stage Assembly Permutation Flowshop Scheduling Problems (DTrSAPFSP) models this operational situation; it is the most recent development in the literature of distributed scheduling problems, which has seen very limited development for possible industrial applications. This research introduces a highly efficient constructive heuristic to contribute to the literature on DTrSAPFSP. Numerical experiments considering a comprehensive set of operational parameters are undertaken to evaluate the performance of the benchmark algorithms. It is shown that the N-list-enhanced Constructive Heuristic algorithm performs significantly better than the current best-performing algorithm and three new metaheuristics in terms of both solution quality and computational time. It can, therefore, be considered a competitive benchmark for future studies on distributed production scheduling and computing.

Solving the Sustainable Automobile Production-Distribution Joint Optimization in the Physical Internet-Enabled Hyperconnected Order-to-Delivery System by I-NSGAIII

The Physical Internet (PI)-enabled hyperconnected order-to-delivery system (OTD) provides new solutions for sustainable supply chains from production perspectives. In this system, a PI-enabled hyperconnected manufacturing system is more closely tied with other functions through Internet-of-Things (IoT)-enabled machines for communication. In the OTD, the PI-enabled hyperconnected production–distribution system (PI-H) is modelled by multi-objective mixed-integer-nonlinear programming to evaluate sustainability. We develop an improved reference-point based non-dominated sorting genetic algorithm (I-NSGAIII) to solve practical-scale PI-enabled hyperconnected production-distribution scheduling problems, with the problem-specific solution expression and dynamic programming, subproblem-guided crossover and mutation strategies, and adaptive evolution mechanisms. I-NSGAIII’s performance advantages and PI-H’s sustainable advantages are validated through extensive experiments.

A network memetic algorithm for energy and labor-aware distributed heterogeneous hybrid flow shop scheduling problem

With the penetration of decentralization into factories, the production scheduling among non-identical factories has emerged as a hot issue for industrial demand response. However, little attention has been paid to differentiated production resources for distributed production scheduling. This paper studies an energy and labor-aware distributed heterogeneous hybrid flow shop scheduling problem (ELDHHFSP). The heterogeneous factors, i.e. different numbers and capabilities of machines, time-of-use electricity price, labor shift, salary and bonus of workers, are included in the ELDHHFSP. The mixed-integer linear programming (MILP) model of ELDHHFSP is presented with the objectives of minimization of total tardiness, total production cost, and total carbon emission. A network memetic algorithm (NMA) is proposed for the ELDHHFSP from the solution and strategy space. The NMA mainly includes two critical parts, i.e., probability network model and learning-based local search, corresponding to global and local search. A probability network model is established to guide the search towards the optimal regions from different directions by employing multiple objective weight vectors. The sampling part samples from the probability matrix of each node in the network to construct new solutions. The learning-based local search integrates several local search operators and a reward learning mechanism is designed to find suitable local search operators when evolutionary. A comprehensive experiment on extensive testing instances is conducted to investigate the effectiveness of the probability network and local-learning strategy. The comparison of NMA and other related algorithms shows the effectiveness and efficiency of NMA.

Optimal spare parts production–distribution scheduling considering operational utility on customer equipment

Integrated scheduling of production and distribution is conducive to saving transportation cost and improving customer satisfaction. However, in previous related literature, no optimization objective has been set to intuitively reflect the influence of the scheduling scheme on the customer side, ignoring the fact that: 1) the production and distribution of spare parts directly affect the performance of operation and maintenance and further affect the operational utility of customer equipment; 2) the delivery time of spare parts has a positive effect on optimizing the operational utility of customer equipment corresponding to the delayed spare parts. To fill this gap, in the paper, we construct an optimal spare parts production scheduling model integrating logistics distribution in flowshop environment, in which equipment operation utility on the customer side is set as an optimization objective and a speed adjustment strategy is introduced, so that the manufacturer can adjust the operation strategy of the equipment corresponding to the delayed spare parts and guide the equipment operation and maintenance in the customer side. There are two objectives in our model. The first is to maximize equipment operational utility on the customer side, and the second is to minimize the total cost on the manufacturing side. To solve the problem, a three-layer hybrid coding memetic algorithm (THMA) is constructed. Two initial heuristics, respectively named Heuristics 1 and Heuristics 2, are constructed. Two local search operators operated respectively on the production and distribution stages of spare parts are constructed. By performing extensive contrast experiments, the superiority of THMA and the effectiveness of all the above operators are verified.

Optimization of integrated production scheduling and vehicle routing problem with batch delivery to multiple customers in supply chain

In supply chain literature, production coordination and vehicle routing have received a lot of attention. Even though all functions in the supply chain are interrelated, they are normally handled independently. This disconnected approach might lead to less-than-ideal outcomes. Increasing total efficiency by integrating manufacturing and delivery scheduling processes is popular. This study focuses on synchronic production–distribution scheduling difficulties, particularly permutation flow shop scheduling in production and sequence-dependent setup time (SDST) and vehicle routing alternatives in distribution. To create a cost-effective distribution among the placement of geographically separated clients and hence to minimize delivery costs, batch delivery to customers employing a succession of homogenized capacity limitation vehicles is examined here. However, this might result in the failure to complete multiple client orders before their deadlines, raising the cost of lateness. As a result, the goal of this study is to lower the overall cost of tardiness and batch distribution in the supply chain. To accomplish so, a mixed-integer nonlinear programming model is developed, and the model is solved using a suggested genetic algorithm (GA). Because there is no established benchmark for this issue, a set of genuine problem scenarios is created in order to assess the proposed GA in a viable and difficult environment. Ruiz's benchmark data, which is derived from Taillard's benchmark cases of permutation flow shops, was supplemented with SDSTs in the production of test examples. In comparison to an exact method, the results show that the proposed GA can rapidly seek solutions to optimality for most small-sized instances. Furthermore, for medium and large-scale cases, the proposed GA continues to work well and produces solutions in a fair amount of time in comparison to GA without the local search.

Multi-local search-based general variable neighborhood search for distributed flow shop scheduling in heterogeneous multi-factories

Heterogeneous multi-factories in different regions bring a challenge to managers in distributed production scheduling. This paper studies a distributed flow shop scheduling problem in heterogeneous multi-factories (DHFSP) with different processing capabilities of machines and electricity prices in multi-factories. The mixed-integer linear programming model (MILP) of the DHFSP is established, where different time-of-use electricity prices and states of machines (running and idle) are integrated. The proposed MILP model is based on position and time-indexed variables where the position of a job in the sequence is modeled and the time horizon is discretized into time slots. Every job on a machine corresponds to a series of position and time-indexed decision variables A multi-local search-based general variable neighborhood search (MGVNS) is presented to address the DHFSP. The earliest due date rule and a distributed version of NEH (Nawaz–Enscore–Ham) are combined to generate a promising initial solution. Four efficient greedy local search methods that move jobs between heterogeneous multi-factories and within the critical factory are presented. The properties of DHFSP are used to reduce the computing time of objectives. A right-shifting procedure is presented to save the electricity cost by considering the total electricity cost consumed by the processing of the jobs and the idle state of the machine between two adjacent jobs. The experiments and investigation are provided on large-sized instances to demonstrate the effectiveness and efficiency of MGVNS. The effectiveness of initialization, accelerated methods, right-shifting procedure, and local search methods are also significant. From the managerial insights, the proposed model can be extended to many real scenarios, such as semiconductor or automobile production, by integrating specific production constraints. The proposed model and scheduling methods provide an economical way to generate efficient schedules to balance the production efficiency and cost.

Integrated production-distribution scheduling with energy considerations for efficient food supply chains

Abstract Quantitative approaches for the integration of production and distribution planning are attracting the interest of scholars and companies in recent years. They can significantly improve supply chain performance and sustainability. In this paper, we propose an optimization model for the integrated scheduling of production and distribution activities, with reference to a real-life company in the food sector. The model takes into consideration changeover times and perishability, and aims to jointly minimize energy, storage and distribution costs. Its applicability is shown through a set of computational experiments, carried out on instances generated from historical data. Two different rescheduling strategies, where the first one reproduces the current behaviour of the firm, are compared. The results show that the current practices of the company can be improved and the model is a valid tool for supporting operational business decisions.

The integrated production-distribution scheduling in parallel machine environment by using improved genetic algorithms

ABSTRACT This study investigates the integrated production-distribution scheduling in a workshop which utilizes the non-identical parallel machines in production line. The objective is to schedule the jobs on machines and forming them into the batches so as to minimize the sum of tardiness and delivery costs. First, a mixed integer linear programming model is proposed which can solve the small-size instances in a reasonable time. To deal with large-size instances, a genetic algorithm (GA) and the improved genetic algorithm (IGA) equipped with optimal properties mechanism is developed. Areal life instance in bedroom furniture manufacturing is also presented that indicates the algorithm can reach the optimal solution for this problem. In order to evaluate the performance of the algorithms, 150 small size instances and more than 800 large size instances are generated which is conducted from benchmarks. The results indicate that IGA outperforms GA with respect to relative percentage deviations.

A memetic algorithm for multi-objective distributed production scheduling: minimizing the makespan and total energy consumption

The classical distributed production scheduling problem (DPSP) assumes that factories are identical, and each factory is composed of just some machines. Inspired by the fact that manufacturers these days typically work across different factories, and each of these factories normally has some workshops, we study an important extension of the DPSP with different factories and workshops (DPFW), where jobs can be processed and transferred between the factories, workshops and machines. To the best of our knowledge, this is the very first time distributed production scheduling with different factories and workshops is studied. We propose a novel memetic algorithm (MA) to solve this DPFW, aiming to minimize the makespan and total energy consumption. The proposed MA is incorporated with a well-designed chromosome encoding method and a balance-transfer initialization method to generate a good initial population. An effective local search operator is also presented to improve the MA’s convergence speed and fully exploit its solution space. A total of 50 DPFW benchmark instances are used to evaluate the performance of our MA. Computational experiments carried out confirm that the MA is able to easily obtain better solutions for the majority of the tested problem instances compared to three other well-known algorithms, demonstrating its superior performance over these algorithms in terms of solution quality. Our proposed method and the results presented here may be helpful for production managers who work with distributed manufacturing systems in scheduling their production activities by considering different factories and workshops. With this DPFW, imbalanced resource loads and unexpected bottlenecks, which regularly arise in traditional DPSP models, can be easily avoided.

Integrated production and transportation on parallel batch machines to minimize total weighted delivery time

This paper considers a production-distribution scheduling problem on parallel batch processing machines (BPMs) with multiple vehicles. In the production stage, the jobs with non-identical sizes and equal processing time are grouped into batches, which are processed on BPMs. In the distribution stage, there are vehicles with identical capacity arriving regularly to transport the batches to the customers. The objective is to minimize the total weighted delivery time of the jobs. A method of computing a lower bound is given to evaluate the proposed algorithms. To tackle this NP-hard problem, a deterministic heuristic (Algorithm H) and two hybrid meta-heuristic algorithms based on ant colony optimization (HACO, MMAS) are proposed, respectively. Through analyzing the property of the investigated problem, the heuristic information and the pheromone trails are defined. Incorporated with a local optimization strategy, the ant colony constructs the schedule first. Then, a heuristic is designed to transport the batches that have been processed. The performance of the proposed algorithms are compared with each other through testing on randomly generated problem instances. It is shown that the proposed MMAS algorithm slightly beats the HACO algorithm, which can find the better solutions than the H algorithm in a reasonable amount of time.

The impact of liner shipping unreliability on the production–distribution scheduling of a decentralized manufacturing system

Liner shipping uncertainty affects global supply chains reliability seriously. The pressure from downstream forces the make-to-order manufacturer to face high penalty cost due to stochastic shipping lead-times. A new risk-averse model is proposed aimed at improving production scheduling reliability integrated with shipment assignments in terms of total operating cost. The closed form of the risk cost of job-shipment assignment under given reliability levels is formulated, which verifies the effects of shipping uncertainty on the production scheduling and operating cost. The computational and statistical evaluation demonstrate our approach can compensate the amplification effects of the high penalty level and shipping uncertainty.

Coordinating order acceptance and integrated production-distribution scheduling with batch delivery considering Third Party Logistics distribution

The present study proposes two mixed integer linear programming model for two important aspects of integrated production-distribution scheduling: order acceptance and batch direct delivery. Moreover, as many companies are unable to provide sufficient transportation facilities to deliver the customer orders due to high costs of initial investment, transportation is outsourced to a third-party logistics provider in which the transportation cost is dependent on the batch. The aim of this paper is trading off among the revenue of accepted orders, costs of delivery, and penalties for tardiness incurred in an integrated production-distribution in a supply chain to maximize the total of benefit. In addition, since the problem in this study is strongly NP-hard, an adaptive genetic algorithm is used to solve large-scale instances in this regard that use the adaptive search approach.A representation procedure is introduced based on two optimal properties of the problem. For the initial population, four heuristics are developed. To explore and locate the algorithm in a better neighborhood, a local search is made use of. Taguchi experimental design was applied to set the appropriate parameters of the algorithms. Moreover, to verify the developed model and evaluate the performance of algorithm against the exact solution, a commercial solver is used. The obtained results on generated random instances reveal the appropriate performance of heuristics, the adaptive approach and local search on the genetic algorithm. Furthermore, the effect of different parameters and factors of the proposed model on the profit shows that the order acceptance and the more vehicles of the company improve the profit.

A Differential Evolution Heuristic for Integrated Production-Distribution Scheduling in Supply Chain Management

A supply chain may be considered as an integrated process in which a group of several organizations, work together. The two core optimization problems in a supply chain are production and distribution planning. In this research, we develop an integrated production-distribution (P-D) model. The problem is formulated as a mixed integer programming (MIP) model, which could then be solved using GAMS optimization software. A differential evolution (DE) algorithm is applied to solve large-sized MIP models. To the best of our knowledge, it is the first paper which applied DE algorithm to solve the integrated (P-D) planning models in supply chain management (SCM). The solutions obtained by GAMS are compared with those obtained from DE and the results show that DE is efficient in terms of computational time and the quality of solutions obtained.

A Hybrid Genetic Algorithm-Simulated Annealing for Integrated Production-Distribution Scheduling in Supply Chain Management

In this paper, we present an integrated production-distribution (P-D) model which considers rail transportation to move deteriorating items. The problem is formulated as a mixed integer programming (MIP) model, which could then be solved using GAMS optimization software. A hybrid genetic algorithm-simulated annealing (GA-SA) is developed to solve the real-size problems in a reasonable time period. The solutions obtained by GAMS are compared with those obtained from the hybrid GA-SA and the results show that the hybrid GA-SA is efficient in terms of computational time and the quality of the solution obtained.

Online integrated production–distribution scheduling problems without preemption

We study an integrated production---distribution scheduling problem where jobs are released by customers to a manufacturer over time. The jobs are released online, that is, at any time the information of the number, release and processing times of future jobs is unknown, and the processing time of a job becomes known when the job is released. The manufacturer processes the jobs on a single machine. During the processing of jobs preemption is not allowed. Completed jobs are delivered in batches to customers via sufficient capacitated vehicles. For the objective of minimizing the sum of the total delivery time and the total distribution cost, we present a 3-competitive algorithm for the single-customer case and then extend the result to the multi-customer case. A lower bound of two on the competitive ratio of the problem is also given.

Aggregate procurement, production, and shipment planning decision problem for a three-echelon supply chain using swarm-based heuristics

Supplier selection is deemed as a crucial strategic decision-making activity in building a competitive edge. Firms prefer to operate with a few trusted suppliers, selected from a bigger pool of vendors. The chosen suppliers are the ones whose commitments are best oriented in realising the business goals of the company. At the same time enterprise targets cannot be achieved in the absence of cost-effective inventory management policies. This has created the inevitable need for aggregate production and distribution planning. Even more competitive strategy would be integrating procurement planning with production-distribution scheduling. We address the problem of integrated procurement, production and shipment planning for a supply chain, spanning over three echelons. Supplier order scheduling is combined with a production-shipment planning process to realise a minimum cost operations policy. Two recently developed swarm heuristics are employed to search for the near optimal solution of the mathematical model, which is developed to capture the aggregate planning problem.

On-line integrated production–distribution scheduling problems with capacitated deliveries

Abstract In on-line integrated production–distribution problems, customers release jobs to a manufacturer that has to process the jobs and deliver them to the customers. The jobs are released on-line, that is, at any time there is no information about future jobs. Processed jobs are grouped into batches, which are delivered to the customers as single shipments. The total cost (to be minimized) is the sum of the total weighted flow time and the total delivery cost. Such on-line integrated production–distribution problems have been studied for the case of uncapacitated batches. We consider the capacitated case with an upper bound on the size of a batch. For several versions of the problem, we present efficient on-line algorithms, and use competitive analysis to study their worst-case performance.

Integrated Order Scheduling and Packing

We consider an integrated production–distribution scheduling model in a make‐to‐order supply chain consisting of one supplier and one customer. The supplier receives a set of orders from the customer at the beginning of a planning horizon. The supplier needs to process all the orders at a single production line, pack the completed orders to form delivery batches, and deliver the batches to the customer. Each order has a weight, and the total weight of the orders packed in a batch must not exceed the capacity of the delivery batch. Each delivery batch incurs a fixed distribution cost. The problem is to find jointly a schedule for order processing and a way of packing completed orders to form delivery batches such that the total distribution cost (or equivalently, the number of delivery batches) is minimized subject to the constraint that a given customer service level is guaranteed. We consider two customer service constraints—meeting the given deadlines of the orders; or requiring the average delivery lead time of the orders to be within a given threshold. Several problems of the model with each of those constraints are considered. We clarify the complexity of each problem and develop fast heuristics for the NP‐hard problems and analyze their worst‐case performance bounds. Our computational results indicate that all the heuristics are capable of generating near optimal solutions quickly for the respective problems.