Table of Content

30 June 2024, Volume 27 Issue 3 Previous Issue    Next Issue

For Selected:

Download Citations EndNote Reference Manager ProCite BibTeX RefWorks

Toggle Thumbnails

Review

A Review of Data-driven Inventory Management Based on Demand Uncertainty SHAO Siqi, ZHONG Yuanguang, CHEN Zhi, LI Yanxi 2024, 27 (3):  1-11.  doi: 10.3969/j.issn.1007-7375.230259 Abstract ( 183 )   HTML ( 27 )   PDF (1318KB) ( 252 )   Save In recent years, with the increasing abundance of high-quality data, continuous development of machine learning techniques and significant improvements of computational capabilities, data-driven inventory management is experiencing unprecedented development opportunities. However, comprehensive and systematic reviews of research advances in this emerging field are currently lacking. In this study, an in-depth analysis of 183 academic papers is conducted using bibliometrics, and the state of the art in this field is visualized through scientific knowledge graphs. Then, the research results of data-driven inventory management from the perspectives of big data and operation management are summarized and synthesized in three aspects: demand information, basic models and basic methods. Essentially, this paper introduces four inventory management models from the perspectives of demand uncertainty and feature data: univariate data-driven newsvendor model, univariate data-driven dynamic inventory model, multi-feature data-driven newsvendor model and multi-feature data-driven dynamic inventory model. On this basis, six main data-driven decision-making methods are summarized: Bayesian analysis, robust optimization, sample average approximation, quantile regression, operation statistics and machine learning. Finally, future research directions and suggestions are discussed from the perspectives of methodologies, tools, challenges, and application hotspots in data-driven inventory management, aiming to provide valuable references and insights for researchers and practitioners in the relevant fields, and to foster the continuous development of data-driven inventory management. References | Related Articles | Metrics

Optimization and Test of Wafer Manufacturing Process

Real-time Scheduling of Wafer Photolithography Area Based on Reinforcement Learning with Gated Recurrent Unit WU Lihui, SHI Jinming, JIN Keshan, ZHANG Jie 2024, 27 (3):  12-21,30.  doi: 10.3969/j.issn.1007-7375.240100 Abstract ( 466 )   HTML ( 17 )   PDF (1331KB) ( 57 )   Save To address the scheduling problem of wafer photolithography area, characterized by dynamic nature, real-time requirements, multiple constraints, and multiple objectives, a real-time scheduling method based on gated recurrent unit (GRU) reinforcement learning is proposed. This method incorporates GRU to learn the temporal information of historical scheduling decisions and states in the photolithography area, providing auxiliary decision-making information for the double deep reinforcement learning (DDRL) model. The input state space and output action set of the DDRL model are designed, and a multi-objective reward function is established with the objective of minimizing the maximum completion time of wafers and maximizing the on-time delivery rate, optimizing the scheduling output by intelligent agents. Additionally, constraint relaxation rules and scheduling methods are proposed combining equipment-specific constraints and mask constraints, to enhance the practicality of scheduling strategies. Through empirical evaluation using real-world cases from a wafer manufacturing enterprise, this method is compared with traditional double deep reinforcement learning and heuristic rule methods for photolithography area, demonstrating its superiority and verifying its effectiveness in solving this problem. References | Related Articles | Metrics

A Rapid Customization Design Method of System-Level Test Equipment for Chips Based on Digital Twins LIN Daqin, ZHAO Rongli, LAI Yuanpeng, LIU Qiang 2024, 27 (3):  22-30.  doi: 10.3969/j.issn.1007-7375.240053 Abstract ( 765 )   HTML ( 15 )   PDF (3116KB) ( 46 )   Save In order to address the issues of long design cycles for system-level test (SLT) equipment caused by high customization requirements and ineffective reuse of design knowledge, a rapid customization design method for SLT equipment is studied. First, the design knowledge prototype of SLT equipment is developed. Additionally, a framework for rapid design of equipment is proposed, including the design knowledge characterization based on core design parameters and a special aggregation method between mechanical modules, i.e., "leaning" behavior, which effectively realizes the storage and reuse of design knowledge for such equipment. Then, based on the developed design knowledge prototype, combined with digital twin technology, we use the digital factory simulation platform developed by our team to encapsulate the SLT equipment component library and build the SLT equipment digital twin design prototype. In this way, the 3D design solutions by parameter configuration can be quickly output. Ultimately, the semi-physical in-the-loop simulation technology is adopted for virtual debugging, making designing, manufacturing, and debugging of the equipment to proceed in parallel and verify each other. Compared with traditional methods, the approach proposed in this paper reduces the cycle time by about 30%, the cost by about 40%, and the labor input by about 60% for designing and debugging of such equipment. References | Related Articles | Metrics

Complex System Modeling & Operation Optimization

Proactive Project Scheduling Optimization of Repetitive Projects ZHOU Guohua, XIA Jing 2024, 27 (3):  31-42.  doi: 10.3969/j.issn.1007-7375.230013 Abstract ( 50 )   HTML ( 17 )   PDF (1560KB) ( 50 )   Save In order to cope with uncertain disturbance encountered in the construction process of repetitive projects, proactive scheduling optimization of repetitive projects is necessary to increase the capacity of project schedules to accommodate uncertainties. In this paper, the float relationships among linear, bar and block activities in repetitive projects is first analyzed based on linear planning. The selection ranges between buffers are determined according to floats. Then, a multi-objective proactive scheduling model is established with the objectives of project duration, cost and robustness. To address its NP-hard nature, a modified particle swarm optimization algorithm (i.e., SA-PSOc) is designed, which incorporate simulated annealing algorithm to avoid particles easily falling into local optima. Algorithm testing shows that the modified algorithm has better global search ability and faster computational speed. Finally, the feasibility and effectiveness of the proposed proactive scheduling optimization model are verified by a case study of a repetitive project. Through simulation, it demonstrates that the proposed method consumes fewer buffers to ensure the project completion on time. Conclusion is made that when preparing a schedule, by increasing resources within a certain range, a better schedule can be achieve. References | Related Articles | Metrics

Modeling and Optimization of Lockage Scheduling for Three Gorges-Gezhouba Dam During Maintenance Periods LIANG Xiaolei, ZHANG Dongmei, ZHANG Yu, HUANG Kai 2024, 27 (3):  43-53,86.  doi: 10.3969/j.issn.1007-7375.220257 Abstract ( 39 )   HTML ( 23 )   PDF (1360KB) ( 59 )   Save To address problems such as long waiting time for ships and low passage efficiency during maintenance periods of the Three Gorges-Gezhouba Dam lock, a lockage scheduling model is developed according to the scheduling requirements of the Three Gorges-Gezhouba Dam lock. The lockage scheduling problem is decomposed into three subproblems: chamber arrangement, lockage assignment and timetable optimization. According to the characteristics of the model, the Bottom-Left algorithm, the lockage assignment algorithm considering navigation equilibrium and the timetable optimization algorithm based on adaptive bald eagle search are established to solve the sub-problems, respectively. In such lockage scheduling problem during maintenance periods, two scenarios including dynamic and static scheduling are analyzed according to the congestion level of ships awaiting lockage. Finally, actual historical data of the navigation from the Three Gorges-Gezhouba dam lock are used for comparing the optimization results of the original and modified bald eagle search algorithms. Results show that in a scheduling cycle, the average waiting time of ships can be reduced by about three hours, the average utilization rate of lock chambers can reach over 75%, and the ship throughput can be increased by six ships. References | Related Articles | Metrics

A Multi-attribute Decision-making Approach Based on Possibility Degree Measure in Interval-Valued q-Rung Orthopair Fuzzy Environment WANG Haolun, LI Xuxiang, XU Tingjun, FENG Liangqing 2024, 27 (3):  54-63.  doi: 10.3969/j.issn.1007-7375.230111 Abstract ( 30 )   HTML ( 21 )   PDF (772KB) ( 37 )   Save Determining the ordering relation of interval-valued q-rung orthopair fuzzy numbers (IVq-ROFNs) is the key to solve the issue of interval-valued q-rung orthopair fuzzy (IVq-ROF) decision-making. In this context, possibility measure is one of the important tools. To address the counter intuitive problem that existing score functions, accuracy functions and possibility measures cannot solve, a multi-attribute decision-making method based on IVq-ROF possibility degree is proposed. Firstly, the possibility measure in IVq-ROF environment is defined, and its related properties are discussed. Then, through some counter intuitive examples, the feasibility and superiority of the IVq-ROF possibility measure are illustrated. Finally, a multi-attribute decision-making model based on IVq-ROF possibility degree is established. Also, two numerical examples are given to analyze the applicability of the existing possibility degree in IVq-ROF environment and verify the effectiveness and rationality of the proposed method. References | Related Articles | Metrics

Tool Wear Prediction Based on Multi-sensor Information Fusion Using Stacked LSTM with Attention Mechanism CHENG Jiawen, BAO Saixialt, ZHANG Chaoyong, LUO Min 2024, 27 (3):  64-77,86.  doi: 10.3969/j.issn.1007-7375.240063 Abstract ( 48 )   HTML ( 26 )   PDF (2675KB) ( 40 )   Save Tool wear is one of the critical factors affecting the quality and efficiency of computer numerical control (CNC) machine tools. To address the current issues of signal singularity and insufficient prediction accuracy in milling cutter wear predictions, a novel method for tool wear prediction is proposed, which involves the fusion of multi-sensor information based on a stacked long short-term memory (LSTM) network with attention mechanism. Initially multiple sensor signals are preprocessed, subsequently multi-domain features are extracted. These features are fused at the feature level using Kernel Principal Component Analysis (KPCA), providing the input for the subsequent network. A stacked LSTM network model with attention mechanism is used to enable adaptive learning of crucial information, achieving a predictive accuracy of 99.9% on the PHM2010 dataset. Comparative experiments are conducted with other algorithms and incorporating artificial noise to verify the high accuracy and robustness of the proposed model. References | Related Articles | Metrics

Intelligent Manufacturing System and Workshop Scheduling Optimization

Distributed Flexible Job Shop Scheduling with Sequence-Dependent Setup Times WANG Youyuan, DONG Bowen 2024, 27 (3):  78-86.  doi: 10.3969/j.issn.1007-7375.230102 Abstract ( 77 )   HTML ( 24 )   PDF (956KB) ( 76 )   Save For the distributed flexible job shop scheduling problem considering sequence-dependent setup times, a mixed-integer linear programming model with the optimization objective of minimizing the makespan is proposed. Also, an improved genetic algorithm is developed. A load-balanced population initialization method is used to improve the quality of the initial population. Six local perturbation operators are constructed according to problem characteristics, and a multiple local perturbation strategy is designed to improve the local search capability of the algorithm. Test cases are generated by extending the flexible job shop scheduling benchmark, and the algorithm parameters are determined by orthogonal experiments. Experimental results show that the proposed strategy can effectively improve the performance of the algorithm, with solutions superior to those obtained by the comparison algorithms, thus verifying the feasibility and effectiveness of the scheduling model and the proposed algorithm. References | Related Articles | Metrics

Equipment Configuration Optimization for an Intelligent Workshop with Multi-resource Collaboration ZHANG Huiyu, LIANG Zhanpeng, WANG Songling, CHEN Qingxin, MAO Ning 2024, 27 (3):  87-97,105.  doi: 10.3969/j.issn.1007-7375.220161 Abstract ( 94 )   HTML ( 16 )   PDF (1964KB) ( 87 )   Save To address the problem of equipment quantity configuration in an intelligent workshop with multi-resource collaboration, an optimization model is established with dual constraints on system output rate and production cycle and the objective of minimizing equipment purchase cost. Since the optimization problem is a stochastic nonlinear integer programming one, and the constraints cannot be expressed in closed form of decision variables, a simulation-based intelligent optimization algorithm is proposed to solve it. For such production workshops with multi-resource collaboration, a system performance evaluation model is developed based on a discrete event simulation platform. A gray wolf optimization algorithm embedded in the simulation model is proposed to obtain the optimization scheme of equipment quantity configuration. Through simulation experiments and optimization case comparisons, the superiority and stability of the proposed method over other algorithms in optimization results are verified. The optimal configuration scheme is determined by analyzing actual application cases, and the results verify the effectiveness of the proposed method, demonstrating its practical value. References | Related Articles | Metrics

An Inventory Control Strategy for MTS/MTO Tandem Production Systems in Stochastic Scenarios LIN Bing, FENG Yi 2024, 27 (3):  98-105.  doi: 10.3969/j.issn.1007-7375.230043 Abstract ( 37 )   HTML ( 13 )   PDF (781KB) ( 28 )   Save This paper investigates inventory control strategies and the importance of production coordination in supply chains. We apply the two-stage tandem production mode of make-to-stock (MTS) and make-to-order (MTO) to the joint production and inventory control in efficient and responsive supply chains. Also, the optimal strategy is characterized. In cases where order arrivals follow a Poisson distribution and production time follows an exponential distribution, through the technique of rate uniformization for Markov process transitions, we formulate a Markov decision process (MDP) model and derive the corresponding Bellman optimality equation. Then, based on the structural properties of the optimal cost function, we characterize the optimal production and inventory control strategies as a basic inventory strategy dependent on system states. Subsequently, the model is extended for cases of batch production. Numerical examples verify the monotone structural property of the optimal strategy and further provide the system performance using the optimal strategy. In addition, we compare the optimal strategy with two other commonly used ones. In the baseline scenario, the capacity limit strategy deviates from the optimal one by 7.36%, while the myopic strategy deviates from the optimal one by 25.73%. In other scenarios, the optimal strategy remarkably outperforms the other two strategies. References | Related Articles | Metrics

A Hybrid Harmony Search Algorithm for Scheduling in a Gas Turbine Manufacturing Workshop LI Minghui, SHI Yuqiang, SHI Xiaoqiu, LI Jia 2024, 27 (3):  106-113.  doi: 10.3969/j.issn.1007-7375.230125 Abstract ( 31 )   HTML ( 14 )   PDF (1066KB) ( 28 )   Save Gas turbine production is a typical kind of discrete manufacturing. The production characteristics of multiple varieties and small batches present challenges to workshop scheduling, resulting in low production efficiency and difficulties of meeting product delivery deadlines. The Harmony Search (HS) algorithm is often used to solve such workshop scheduling problems due to its simplicity and ease of operation. However, the convergence rate of traditional HS algorithm is relatively low, and it is easy to get trapped in local optima. Accordingly, this paper builds a mathematical model for scheduling in a gas turbine manufacturing workshop with the objective of minimizing the maximum completion time. A discrete improved multi-population hybrid HS algorithm is proposed to solve the problem. Combining the advantages of HS algorithm and the variable neighborhood search algorithm, we propose an encoding method based on operations. The Metropolis rule of simulated annealing is used in population iteration to improve population diversity. An adaptive memory retention probability and pitch adjusting rate are proposed to adjust parameters, improving the global optimization capability of the algorithm. We also incorporate variable neighborhood searching to accelerate the convergence of the proposed algorithm. Performance tests and case studies show that the proposed algorithm outperforms existing algorithms. References | Related Articles | Metrics

Intelligent Optimization of Wind Turbine Extrusion Plates Production Scheduling Using a Hybrid Multi-strategy Multi-objective Heuristic Sparrow Search Algorithm ZHANG Zhiwei, LI Luoping, YANG Xiaoying, YANG Xin 2024, 27 (3):  114-129.  doi: 10.3969/j.issn.1007-7375.240042 Abstract ( 55 )   HTML ( 13 )   PDF (3690KB) ( 32 )   Save In order to address the production scheduling problem of wind turbine extrusion plates (PSP-WTEP) with sequence-dependent adjustment time and sequential alignment constraints, a multi-objective optimization model is developed to minimize equipment load deviation, delivery time deviation and maximize equipment utilization. A modified multi-objective heuristic sparrow search algorithm (MHSSA) is designed based on the mechanisms of Pareto optimization and crowding distance calculation. A two-layer encoding strategy of "component-region" and a heuristic decoding operator of "inversion-repair-optimization" are incorporated in the algorithm. An improved population initialization strategy, which combines multiple rules with opposition-based learning, is adopted to enhance the global search capability of the algorithm. Additionally, an improved search strategy with crossover operators and an external archive disturbance mechanism is utilized to enhance the optimization accuracy and population diversity of the algorithm. The effectiveness of the proposed optimization model and intelligent scheduling algorithm is verified through dataset testing and instance simulation analysis. References | Related Articles | Metrics

Sustainable Operation and Supply Chain Management

A Study on Recycler Production Modes Considering Brand Advantages: Remanufacturing or Refurbishing GUO Qiang, ZHANG Congcong, NIE Jiajia 2024, 27 (3):  130-137,158.  doi: 10.3969/j.issn.1007-7375.230080 Abstract ( 356 )   HTML ( 15 )   PDF (1098KB) ( 47 )   Save This paper investigates the influence mechanism of brand advantages on the choice of production modes for recyclers. Based on a tripartite game model, the decision conditions and benefit distribution results of recyclers choosing between remanufacturing and refurbishing with different brand advantages are analyzed using methods such as KKT conditions. Results indicate that with a low brand advantage, the choice of production modes for recyclers depends on recycling cost: low recycling cost leads to full refurbishment as the optimal strategy for recyclers, but it harms the interests of brand manufacturers and brand retailers; moderate recycling cost results in partial refurbishment for recyclers, where brand retailers benefit, but brand manufacturers are impacted; high recycling cost reduces the refurbishment threat of recyclers, and both brand manufacturers and brand retailers accept this outcome by default. With a high brand advantage, recyclers and brand manufacturers are motivated to reach a remanufacturing agreement, achieving a win-win situation for both parties, while brand retailers oppose this agreement strongly. Moreover, cooperative remanufacturing mode may not necessarily lead to high consumer surplus. In certain circumstances, consumer surplus in the refurbishment competition mode may exceed that in the cooperative remanufacturing mode. References | Related Articles | Metrics

Channel Selection Considering the Sensitivity of Carbon Emission and Delivery Time WU Xiaojie, LIU Xin, LIN Xiaogang, LIU Huan 2024, 27 (3):  138-146.  doi: 10.3969/j.issn.1007-7375.230041 Abstract ( 365 )   HTML ( 17 )   PDF (838KB) ( 48 )   Save To study the strategy selection of manufacturers between traditional single retail and online retail channels in a market environment where consumers are sensitive to both carbon emission and delivery time, this paper constructs two supply chain structures: a traditional single retail channel and a dual channel of online direct sales, composed of manufacturers and traditional retailers. The equilibrium pricing strategies and the optimal profits of supply chain members of the two models are compared and analyzed, exploring channel preferences between retailers and manufacturers in the context of consumer dual sensitivity. Results indicate that the carbon reduction level of dual-channels is always higher than that of single-channels. When consumers prefer online direct channels and both consumer dual sensitivity and channel competition in dual channels are low, retailers are willing to accept a higher wholesale price and prefer traditional single retail channels; otherwise, they prefer dual channels of online direct sales. When consumers prefer online direct channels, manufacturers prefer the dual-channel model of online direct sales; however, when consumers do not prefer online direct channels and both consumer dual sensitivity and channel competition in dual channels are low, manufacturers prefer the model of traditional single-retail channels; otherwise, even if consumers do not prefer online channels, manufacturers prefer dual channels of online direct sales when either consumer dual sensitivity or competition is high. Numerical analysis further confirms the conclusions of this paper. References | Related Articles | Metrics

Network Equilibrium of Fresh Supply Chains Considering Effort Levels of All Parties with the Goal of Carbon Peak and Carbon Neutralization GUO Wenqiang, LIANG Yunze, GUO Jinzhong 2024, 27 (3):  147-158.  doi: 10.3969/j.issn.1007-7375.240027 Abstract ( 333 )   HTML ( 17 )   PDF (833KB) ( 33 )   Save In order to explore the optimal effort levels of fresh supply chain members under different competition and cooperation modes, as well as the impact of consumer low-carbon preferences and raw material capacity constraints on supply chain profits, this study, considering the perishable and seasonal nature of fresh products, focuses on a fresh supply chain comprising multiple suppliers, manufacturers, retailers, and demand markets. The network equilibrium method is adopted to develop a supply chain network model and design an algorithm based on logarithmic quadratic approximation for solving the model. Results show that when all members of a fresh supply chain exert collaborate efforts, avoiding excessive effort investment can optimize profits of relevant members and the overall supply chain. When efforts are coordinated among different levels of a fresh supply chain, the improvement in the efforts of suppliers and retailers is more conducive to maximizing overall profits. When members of a fresh supply chain work independently, decision-makers need to pay close attention to the effort levels of competitors or other hierarchical members. The increase in consumer low-carbon preferences benefits manufacturer profits, but has limited improvement in the profits of suppliers and retailers. Raw material capacity constraints can significantly improve supplier profits, but are not conducive to improving profits for other supply chain members. References | Related Articles | Metrics

A Quality Information Disclosure Strategy for Dual-Channel Supply Chain Products with Manufacturer Dominance OU Jian, CHANG Yuan, MIN Jie, CAO Zonghong 2024, 27 (3):  159-170.  doi: 10.3969/j.issn.1007-7375.230023 Abstract ( 51 )   HTML ( 18 )   PDF (1602KB) ( 62 )   Save In a dual-channel supply chain composed of manufacturers and platform retailers, considering the asymmetric cross-channel influence of quality information disclosure between manufacturers and platform retailers, a disclosure strategy with manufacturer dominance and a game model of channel pricing are established. The pricing behavior of manufacturers and platform retailers is studied with the non-disclosure strategy, the disclosure strategy with manufacturer dominance and the disclosure strategy with retailer dominance. The quality disclosure thresholds for manufacturers and retailers are determined through comparison, while the impact of product quality levels and the cross-channel influence of platform retailer disclosure on the game equilibrium is analyzed. Results show that: 1) a high quality level induces manufacturers to proactively disclose quality information, while a low quality level makes manufacturers more inclined to retailer disclosure; 2) the disclosure willingness of platform retailers increases with the improvement of their cross-channel influence; 3) when the quality level is high and retailers have significant cross-channel influence, both manufacturers and retailers prefer retailer disclosure; 4) high cross-channel influence of disclosure is not always beneficial to platform retailers: when the cross-channel influence of retailers exceeds a certain threshold, the profit of retailers decreases significantly due to the change of disclosure strategy first, and then increases as retailer influence grows. References | Related Articles | Metrics