International Journal of Industrial Engineering: Theory, Applications and Practice
Not a member yet
943 research outputs found
Sort by
Joint Optimization of Vehicle and Driver Scheduling for Pure Electric Bus Line
Get PDFPure electric buses are rapidly being promoted in various cities. However, the unbalanced distribution of tasks and underutilization of vehicles and drivers is a growing problem. This paper proposes an optimization model to solve these problems while considering factors such as vehicle charging demand and driver working hours. The objective of the model is to minimize the fleet size and number of drivers, aiming to strategically schedule the rest time for drivers and charging time for vehicles within the same period of time. The model involves an improved shifting trip departure time algorithm for charging and resting time windows and an algorithm for generating vehicle and driver chains. Finally, the proposed model and algorithms are applied to bus route no. 31 in Harbin. The results indicate that reasonable scheduling plans can reduce the fleet size and number of drivers while increasing the average utilization rates of buses and drivers
Research On Trajectory Planning Control of Industrial Manipulator Based on ALO Algorithm
Get PDF
Aiming at the shortcomings of the ant lion optimization algorithm (ALO) in industrial manipulator trajectory planning, such as long path length, time-consuming rotation time, and uneven path, an improved ALO (IALO) is proposed. Firstly, the population is initialized by cubic chaotic mapping to improve the quality of ant lion population. Secondly, the trust region mutation is used to improve the location update mode of ant lion population and balance the global search ability and local mining ability. Finally, the Gaussian mutation disturbance strategy is used to improve the location update mode of ant lion population and enhance the ability of the algorithm to jump out of local optimization. Taking trajectory length, rotation time, and redundancy rate as indicators, compared with the ABC algorithm and classic ALO, this algorithm has a shorter path length and less rotation time
Integrating The Entropy and Picture Fuzzy Set Methods to Process Supplier Selection Issues
No full textMany factors must be considered when selecting suitable and stable suppliers. It directly affects the success and sustainable development of supply chains, is a complex multi-criteria decision-making (MCDM) problem, and is a core issue for supply chains. Due to the COVID-19 outbreak, the supply and demand of supply chains worldwide have been severely imbalanced. The overall economic environment remains uncertain after the epidemic, so it is challenging for experts to effectively apply traditional MCDM research methods to evaluate and select suppliers. Therefore, this study integrates the entropy and picture fuzzy set methods and the VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) approach to process supplier selection issues. In the numerical verification, this study uses aquatic product e-commerce for supplier selection as an illustrative example to further verify the effectiveness and applicability of the proposed methods. The calculation results were compared with the typical fuzzy set, the intuitionistic fuzzy set, and the original picture fuzzy set. The numerical verification results comparing the different listed calculation methods show that the proposed method can overcome the unclear, missing, or incomplete evaluation information and effectively handle the MCDM problem in the fuzzy environment
A Novel Hybrid Method for Intelligent Machining Feature Recognition in Manufacturing Systems
No full textA matter-element and graph-based machining feature recognition method is proposed to address the recognition of complex machining features and provide corresponding tool adaptation interfaces. First, since complex features are formed through Boolean operations of basic features, an inclusion relationship necessarily exists between complex and basic features. Therefore, the matter-element model, which excels at representing inter-object relationships, is employed to describe complex features, basic features, and their relationships. Next, an Attributed Adjacency Graph (AAG)-based algorithm is introduced to decompose the entire part and derive AAGs of machining features. The corresponding Attribute Adjacency Matrix (AAM) for each machining feature is constructed based on geometric element coding rules to enable basic feature recognition. Furthermore, using shared surfaces and edges extracted from the STEP neutral file, the recognized basic features are systematically organized as matter-element structures to represent complex features. The critical dimensions of complex features are determined by comparing the geometric dimensions of the constituent basic features. Finally, a platform developed in Java 1.8 demonstrates the method’s practicality through a case study. Results indicate that the proposed method is not only straightforward to implement but also readily integrable with cutting processes
Equity-Oriented Two-Echelon Vehicle Routing Problem: A Three-Phase Heuristic Algorithm
No full textFierce competition and the requirement for sustainable development compel catering services and urban logistics industries to balance cost-efficient transportation with improved service quality and customer equity. The two-echelon vehicle cooperation system, where a primary vehicle (truck) serves as a mobile base for a secondary vehicle (UAV), has gained attention for its potential to leverage the strengths of both vehicle types, enhancing operational efficiency and service delivery. This paper presents an equity-oriented two-echelon vehicle operation problem, where trucks and UAVs cooperate to provide equitable services. We model the problem as a mixed-integer linear program (MILP), incorporating equity considerations through a set of constraints. Specifically, we adopt the relative range scheme from the literature as an equity indicator, aiming to minimize the relative deviation between the maximum and minimum arrival times for unit demand across customers. To solve it, we propose a three-phase heuristic algorithm that dynamically adjusts equity constraints while minimizing transportation costs. Numerical experiments across various instance sizes show that the algorithm consistently produces high-quality solutions with optimality gaps of less than 10%
Smart Manufacturing Systems Under Transportation and Energy Management Constraints
Get PDFIn the Industry 4.0 era, achieving high energy efficiency and flexibility in manufacturing systems remains a critical challenge, particularly in the integration of Automated Guided Vehicles with on-board batteries. This study addresses this gap by introducing a novel evolutionary approach that simultaneously optimizes task scheduling, transport operations, and vehicle battery management—pioneering a comprehensive solution for manufacturing efficiency. Unlike traditional methods, our approach not only reduces lead times and operational costs but also extends battery lifespan through improved energy management strategies. Experimental results demonstrate significant advancements, including a 29.25% improvement in battery levels and a 3.64% reduction in production time. These results surpass established benchmarks in 88.23% of test cases. These outcomes not only enhance sustainability and operational resilience but also provide actionable insights for implementing more efficient and competitive Industry 4.0 manufacturing systems. By addressing critical challenges in energy and operational management, this study lays the groundwork for future innovations in sustainable and adaptive manufacturing practices
Simulation Model of Container Terminal Logistics System Based on Multi-Scale Information Theory
No full textThe logistics system of container terminals involves many transport resources and processes, and simulation models can help decision makers optimize management systems and improve decision-making. To improve the productivity of container terminals, the shortcomings of scale optimization and static design of the traditional container terminal logistics system are improved. The study constructs a simulation model of a logistics system based on multi-scale information theory and introduces a queuing network to analyze the internal logic characteristics of different parts of the system. Meanwhile, an improved bat algorithm is introduced to design the dynamic scheduling scheme for the berth scheduling problem in the container terminal logistics system. The simulation results showed that the improved algorithm tended to be stable when the number of iterations reached 160 times. The allocation under the improved berth scheduling system showed good uniformity in time and space, and the cost of unit deviation was less than 3,000 RMB. The container terminal logistics system designed in this study can improve logistics efficiency, reduce transportation costs, and meet the transportation needs of loading and unloading ships and vehicles
Intelligent Network Orchestration System for Efficient Deployment and Management of Distributed Deep Neural Networks in Dynamic Network Environments
Get PDFThe evolution of distributed deep neural networks (DNNs) in device-edge-cloud architectures necessitates adaptive network control mechanisms to meet stringent latency requirements. Existing software-defined networking (SDN) solutions exhibit limitations in dynamic resource orchestration and cross-layer optimization for AI workloads. We present INOS (Intelligent Network Orchestration System), an SDN-based framework integrating network function virtualization to enable QoS-aware network slicing for prioritized DNN traffic and deep reinforcement learning-driven task offloading across heterogeneous compute nodes. Through NS-3 simulations replicating industrial IoT scenarios, INOS demonstrates quantifiable improvements in latency reduction and resource efficiency compared to static resource allocation baselines. The system architecture extends SDN control plane capabilities with AI-native decision modules, addressing key challenges in service function chaining for distributed intelligence
Impact of COVID-19 on The Bullwhip Effect Across U.S. Industries
No full textThe Bullwhip Effect, describing the amplification of demand variability up the supply chain, poses significant challenges in Supply Chain Management. This study examines how the COVID-19 pandemic intensified the Bullwhip Effect across U.S. industries, using extensive industry-level data. By focusing on the manufacturing, retailer, and wholesaler sectors, the research explores how external shocks exacerbate this phenomenon. Employing both traditional and advanced empirical techniques, the analysis reveals that COVID-19 significantly amplified the Bullwhip Effect, with industries displaying varied responses to the same external shock. These differences suggest that supply chain structures play a critical role in either mitigating or intensifying the effect. By analyzing the dynamics during the pandemic, this study provides valuable insights into managing supply chains under global disruptions and highlights the importance of tailoring strategies to industry-specific characteristics
Self-Live or KOL-Live? Selling Channel Selection in A Platform Supply Chain Considering Return Policies
Get PDFLive-streaming sales in e-commerce have become increasingly indispensable. This paper examines the factors influencing supply chain members’ choice between self-run (channel S) and KOL live streaming (channel K) in live commerce and investigates how different return policies—return refund (RR) and returnless refund (RL)—affect the platform’s channel selection. It shows that brands prefer the live-streaming channel under both return policies. Under return refund, the platform favors live streaming when the KOL’s influence is weak or strong, but avoids self-run live streaming when the KOL’s influence is moderate. Under the returnless refund, the platform chooses self-run live streaming when the KOL's influence is moderate, and the proportion of opportunists is low. However, as the proportion of opportunists rises, the platform avoids self-run live streaming. Only when the KOL's influence is negligible, and the proportion of opportunists is high, will the brand choose to return a refund