Table of Content

30 April 2024, Volume 27 Issue 2 Previous Issue    Next Issue

For Selected:

Download Citations EndNote Reference Manager ProCite BibTeX RefWorks

Toggle Thumbnails

Operational Management in the Context of Industrial Internet of Things: A Review WANG Kangzhou, WANG Dongdong, DOU Lei, XUE Lin 2024, 27 (2):  1-13.  doi: 10.3969/j.issn.1007-7375.230247 Abstract ( 621 )   HTML ( 81 )   PDF (707KB) ( 464 )   Save The Industrial Internet of Things (IIoT), as a new industrial ecosystem integrating advanced information technologies with manufacturing, holds significant importance for enhancing operational efficiency and promoting high-quality development in the manufacturing sector. A comprehensive analysis of literature related to operational management in the IIoT context reveals several key findings. 1) Research on IIoT platforms often employs qualitative methods to analyze their application scenarios and ecosystems across various industries. However, it is essential to focus on quantitative methods to explore operational and coordination mechanisms in depth. 2) Existing studies on value creation suggest enhancing product quality, reducing costs, and optimizing business processes through digitalization, networking, and intelligence. Thereby, it highlights the importance of understanding value co-creation mechanisms and patterns among multiple stakeholders, including manufacturing enterprises and customers. 3) Current research primarily concentrates on optimizing individual activities in production operations using IIoT technologies, mathematical models and algorithms. Nevertheless, there exists a critical need to investigate methods for constructing integrated collaborative processes with the cycle of “research, manufacturing and maintenance” within the IIoT environment, along with exploring multi-level closed-loop decision-making systems and intelligent decision-making methods. 4) It is demonstrated that supply chain management in IIoT can improve operational performance through interconnection, visibility, real-time capabilities, and traceability, while the impact of technology adoption on strategy selection and coordination mechanisms is also explored. Future attention may shift towards downstream clients in supply chain management. References | Related Articles | Metrics

A Review of Trustworthy Machine Learning CHEN Caihua, SHE Chengxi, WANG Qingyang 2024, 27 (2):  14-26.  doi: 10.3969/j.issn.1007-7375.230241 Abstract ( 468 )   HTML ( 45 )   PDF (937KB) ( 624 )   Save Machine learning technology is continuously evolving and is extensively applied across various domains, demonstrating capabilities beyond human abilities. However, improper use of machine learning methods or biased decision-making can harm human interests, especially in sensitive areas with high-security demand such as finance and healthcare, etc., leading to an increasing attention on the trustworthiness of machine learning. Currently, machine learning technology commonly exhibits several drawbacks, such as biases against underrepresented groups, lack of user privacy protection, lack of model interpretability, and vulnerability to threats and attacks. These shortcomings undermine human trust in machine learning methods. Although researchers have conducted targeted studies on these issues, there is a lack of a comprehensive framework and methodology to systematically provide trustworthy analysis of machine learning. Therefore, this paper reviews the current mainstream definitions, indicators, methods, and evaluations of fairness, interpretability, robustness, and privacy in machine learning. Then, the relationships among these elements are discussed, while a trustworthy machine learning framework is established by integrating an entire lifecycle of machine learning. Finally, we present some of the current issues and challenges awaiting resolution in the field of trustworthy machine learning. References | Related Articles | Metrics

A Review on Surface Defect Detection Based on Deep Intelligent Vision GAO Yiping, WANG Hao, LI Xinyu, GAO Liang 2024, 27 (2):  27-36,66.  doi: 10.3969/j.issn.1007-7375.230233 Abstract ( 557 )   HTML ( 54 )   PDF (1036KB) ( 493 )   Save The exploration on surface defect detection based on deep intelligent vision plays an increasingly important role in the manufacturing industry. The importance of surface defect detection based on deep intelligent vision in modern industrial quality inspection is explained and the existing research progress is summarized in this paper. Deep intelligent vision provides high-precision and high-efficiency surface defect detection algorithms for different industrial scenarios based on the technologies of machine vision and deep learning. Surface defect detection can be divided into three categories: surface defect classification, localization, and segmentation from the perspective of detection fineness. The classification, localization, and segmentation methods are systematically reviewed, respectively, to sort out the problematic points and lines of the existing surface defect detection methods. Surface defect classification focuses on the problem of data and defective graphical features, which shows decentralized development due to its basic and easily expandable nature for application in different industrial scenarios. Surface defect localization takes the model framework, rectangular box detection mechanism, and annotation cost as the main problems, showing a research trend of pursuing lightweight and feature fusion mechanisms. Surface defect segmentation pays more attention to detailed features of an image. A multi-task framework for classification, localization, and segmentation, is studied to explore the complementarity between classification and segmentation detection. Finally, the current issues of existing surface defect detection studies are concluded and an outlook on the development trend is given. References | Related Articles | Metrics

Human Factors Engineering

Three-dimensional Bin Packing Considering Working Posture Comfort XU Xiangbin, YU Qinfang 2024, 27 (2):  37-47.  doi: 10.3969/j.issn.1007-7375.230139 Abstract ( 278 )   HTML ( 34 )   PDF (1690KB) ( 161 )   Save This paper aims to reduce the probability of stevedores suffering from musculoskeletal diseases due to repeated bending during the cargo loading process, optimize their work posture comfort to reduce work fatigue, and thus improve the overall benefits of society. Based on the perspective of co-optimization of human factors engineering and operational research, we propose and study the vehicle loading problem that consider working posture comfort from the perspective of comfortable postures for stevedores. Firstly, the comfort of the loading work posture is evaluated to establish the model. Secondly, key issues such as cargo sorting optimization and cargo placement rules are studied. An algorithm combining the maximal-space heuristics and biased random-key genetic algorithm is designed. Finally, experimental verification is conducted through instances. Results show that the proposed model and algorithm can improve the comfort of stevedores' work postures without increasing vehicle transportation cost, Moreover, there is a greater optimization space of the work posture comfort for cargoes with relatively small sizes and scales, verifying the effectiveness of the model and algorithm. References | Related Articles | Metrics

Effects of Colour Contrast and Information Density on Visual Search by Older Users QIN Hua, GUO Bowen, WANG Zhongting, RAN Linghua, CHEN Yongquan, ZOU Chuanyu, HE Yue 2024, 27 (2):  48-56,86.  doi: 10.3969/j.issn.1007-7375.230197 Abstract ( 262 )   HTML ( 30 )   PDF (1864KB) ( 234 )   Save The growing use of dynamic interfaces for presenting medical information in hospitals has led to challenges for the elders with declining visual function, who frequently face difficulties in accessing and identifying dynamic directional triage information within healthcare settings. In order to improve the performance of elderly users in observing dynamically oriented information interfaces, a study is conducted on the factors of color contrast and information density that significantly affect the visual search performance of elderly users. Based on the hospital dynamic triage display interface and the E-prime experimental platform, visual search experiments are conducted on twenty elderly users (over 60 years old) by using color contrast and information density as independent variables. The data of visual search time, visual search correctness, clarity, and comfort ratings are collected at different levels of independent variables. Results show that the color contrast ratio of 13.7∶1 has the best visual search performance and evaluation of elderly users. However, when the value rises to 17.4∶1 and 21∶1, the performance and evaluation significantly decrease, but still better than the ratio of 8.7∶1. For the chosen font and display conditions, elderly users showed significantly better visual search performance and evaluation results at the 0.46 and 0.55 (5-6 lines) information density levels compared with others levels. Furthermore, the 0.74 (8 lines) information density level is found to be the maximum suitable density for elderly users. References | Related Articles | Metrics

An Evaluation Method for Intelligent Voice Interaction Design Based on User Perceptual Preference WANG Tiedan, HU Yipeng, PENG Dinghong 2024, 27 (2):  57-66.  doi: 10.3969/j.issn.1007-7375.230258 Abstract ( 279 )   HTML ( 29 )   PDF (838KB) ( 190 )   Save User perceptual preference is an important basis for intelligent voice interaction design. To solve the problem of inconsistency between users' actual decision-making behavior and their perceptual evaluation results in the evaluation process, an intuitionistic fuzzy anchoring evaluation method based on user perceptual preference is proposed. Firstly, to fully describe the ambiguity and uncertainty of user preference, the intuitionistic fuzzy set is used to represent the user's perceptual preference information for intelligent voice interaction design. Secondly, the best and worst idea is incorporated into the ordinal priority approach (OPA) to determine the weights, which ensures the simplicity of the operation process and overcomes the defects of traditional OPA in sorting. Thirdly, the mechanism of anchoring effect in the evaluation process is described from a mathematical perspective, and the anchoring effect is integrated into the evaluation method of intelligent voice interaction design to quantitatively analyze the impact of the anchoring effect on the final evaluation result. Finally, taking the voice interaction design of an intelligent vehicle system as an example, it is verified that the method can improve the accuracy and reliability of predicting the actual decision result of users. References | Related Articles | Metrics

A Synthetic Evaluation Method for Mental Workload in Complex Assembly Task YU Qiwei, TANG Weihao, GENG Jie 2024, 27 (2):  67-73,97.  doi: 10.3969/j.issn.1007-7375.230183 Abstract ( 260 )   HTML ( 34 )   PDF (1205KB) ( 193 )   Save Excessive mental workload in complex assembly operations can reduce assembly performance and quality, and may even lead to safety accidents. Therefore, accurately measuring and evaluating the mental workload of assembly tasks to determine the level of mental workload for operators is an important basis for optimizing production systems. In order to evaluate the mental workload in complex assembly tasks, assembly task experiments based on Lego block simulation are designed. Data from 25 indicators are collected by subjective measurement, performance measurement and physiological measurement to analyze the sensitivity of each indicator to changes of mental workload. Results show that seven indicators comprised of the performance measurement indicator, subjective measurement indicator, fixation times, total fixation duration, scanning times, total scanning duration and average pupil diameter are the effective measurement ones for the mental workload in complex assembly tasks. Based on these effective measurement indicators, a synthetic evaluation model for assembly task mental workload is established by two modeling methods, namely BP neural network and Bayesian linear discrimination. Findings show that the BP neural network model, which uses factor analysis to transform the 7-D indicators into the 2-D synthetic indicators as the principal component input, and employs the normalized conjugate gradient method as the training algorithm, is the optimal synthetic evaluation model for mental workload in assembly tasks, with the accuracy of 84.80%. The proposed method for measuring and evaluating mental workload can provide reference for the evaluation and optimization of mental workload in assembly operations. References | Related Articles | Metrics

Industrial Interconnection & Manufacturing Service Management

User Profiles and Modeling for Iterative Product Designing with Sentiment Analysis and Data-driven Approaches ZHOU Yanjie, LI Yaohui, WANG Yu, WANG Yongsheng 2024, 27 (2):  74-86.  doi: 10.3969/j.issn.1007-7375.230227 Abstract ( 259 )   HTML ( 25 )   PDF (1755KB) ( 186 )   Save To address the issues of time-consuming and laborious, as well as low efficiency of traditional product designing, user profiles and a modeling method for iterative product designing is proposed. The proposed method can effectively capture use attitudes towards product features, providing a new baseline for enterprises to conduct product iteration. First, the user information attributes and online user comments are obtained by web crawler technology. A conceptual model of three-dimensional user profiles is established by integrating product information attributes. Based on product features, a profile model is developed using Word2vec technology and sentiment analysis from two dimensions: user attention and quality satisfaction. Then, by analyzing the overall user profile with the fine-grained feature profile of products, a product optimization strategy is determined using the K-value method, emphasizing both core advantageous features and features for optimization. Finally, a case study is conducted using a certain camera as an example to analyze data-driven iterative product designing. Results show that the proposed method can effectively explore the core demand of users and provide enhanced product design solutions for enterprises to expedite product iterations. References | Related Articles | Metrics

Automatic Designing of Scheduling Rules for Heterogenous Parallel Machines with Setup Time ZHONG Hongyang, LIU Jianjun, ZENG Chuangfeng, CHEN Qingxin, MAO Ning 2024, 27 (2):  87-97.  doi: 10.3969/j.issn.1007-7375.220061 Abstract ( 225 )   HTML ( 37 )   PDF (1213KB) ( 144 )   Save Taking massive customized production of home appliance as the research background, the scheduling of home appliance manufacturing is abstracted as a problem of dynamic heterogenous parallel machine scheduling with sequence-dependent setup time (HPMS-SST). Manual scheduling rules are simple and efficient in solving dynamic scheduling problems, but their adaptability to different scenarios is weak. To this end, an automatic design framework for rules based on genetic programming (GP) is introduced. First, by analyzing the features and optimization requirements of home appliance production, a model of HPMS-SST is established with the objective of minimizing the mean product tardiness. Subsequently, based on the characteristics of this problem, an improved GP algorithm is proposed for the coevolution of machine assignment and queue sequencing rules; Also, the feature attributes of machines and orders are extracted and input into the GP algorithm framework to automatically design scheduling rules; Finally, a number of test cases are generated from the real production data of a home appliance manufacturer. By comparing the experimental results of the proposed algorithm and manual designed rules in various conditions, the effectiveness of the GP algorithm is verified. Besides, sensitive analysis is conducted to evaluate the influence of parameters for different production conditions on the generated GP-based rules. References | Related Articles | Metrics

Quality Prediction for Small-batch Production Based on SMOTE-IKPCA-SeNet Deep Transfer Learning YANG Jianfeng, CUI Shaohong, DUAN Jiaqi, WANG Ning 2024, 27 (2):  98-106,157.  doi: 10.3969/j.issn.1007-7375.230121 Abstract ( 222 )   HTML ( 26 )   PDF (1281KB) ( 165 )   Save With the development of intelligent manufacturing technology and the growing demand for personalization, multi-variety and small-batch production has gradually become the mainstream in the manufacturing industry. In this condition, traditional quality management methods which focus on large-batch production and statistical process control are not suitable for small-batch production. In complex production processes, it also exists challenges such as numerous parameters, non-linearity and interactions. To this end, a deep transfer learning method is adopted to predict the quality of target products using small sample data transferred from massive historical production data. First, by using the synthetic minority over-sampling technique (SMOTE) algorithm and an improved kernel principal component analysis (KPCA) algorithm, transferable features from both the source and target domains are selected, balancing feature importance and transferability. It also mitigates negative transfer issues and enhances the generalization capability of the model. Then, a quality prediction model based on deep transfer learning is built using a convolutional neural network, i.e., SeNet, which incorporates a channel attention mechanism. Simulation results demonstrate that as the number of target domain samples increases, the proposed method is significantly superior to prediction accuracy compared with the widely adopted support vector machine modeling method. Additionally, the proposed selection method of transferable features significantly enhances the quality prediction performance of deep transfer learning, providing a novel approach to ensuring the quality of complex small-batch production processes. References | Related Articles | Metrics

Order Dispatching and Routing for Decentralized Joint Distribution Considering Empty-loading Losses ZHANG Meng, SUN Lulu, SU Bing, WANG Nengmin 2024, 27 (2):  107-118,137.  doi: 10.3969/j.issn.1007-7375.230228 Abstract ( 227 )   HTML ( 31 )   PDF (1624KB) ( 124 )   Save The high empty-loading rate in logistics activities may result from unreasonable routing and insufficient cooperation among enterprises. Joint distribution is an effective mode to reduce empty-loading losses. However, under the condition of decentralized joint distribution, logistics enterprises may choose the routes with minimum cost based on the assigned orders, resulting in an increase in empty-loading losses of the joint distribution alliance. This study investigates order dispatching and routing for decentralized joint distribution considering empty-loading losses. A definition of empty-loading losses is given firstly. Then, with the trade-off between the objectives of minimizing cost and empty-loading losses in the entire distribution process, an order dispatching strategy is proposed based on the characteristics of decentralized joint distribution mode, and the order dispatching and routing optimization models are developed. A precise algorithm based on epsilon constraint method, an improved MOPSO (Multiple Objective Particle Swarm Optimization) algorithm, and a polynomial-time fast algorithm are developed to solve the problem. The effectiveness of the proposed algorithms is verified based on numerical instances. Analysis results indicate that even if logistics companies pursue cost minimization, the proposed order dispatching strategy can achieve results similar to global optimization. References | Related Articles | Metrics

Optimization of AGV Configuration in a Flexible Job Shop with Random Batch Transportation ZHANG Huiyu, WANG Songling, CHEN Qingxin, MAO Ning 2024, 27 (2):  119-129.  doi: 10.3969/j.issn.1007-7375.220110 Abstract ( 246 )   HTML ( 26 )   PDF (1217KB) ( 180 )   Save To solve the issue of configuring AGV quantity for a material handling system in a flexible job shop, an optimization model with dual constraints of system output rate and production cycle is established to minimize the AGV purchase cost. Since the optimization problem is a stochastic nonlinear integer programming one, and the constraints cannot be expressed in a closed form of decision variables, a simulation based on particle swarm optimization algorithm is proposed to solve it. For a flexible job shop with random batch transportation, a performance estimation model is established based on a discrete event simulation platform, and a particle swarm optimization algorithm embedded in the simulation model is proposed to generate the optimization scheme of AGV quantity configuration. Through simulation experiments and comparison of different optimization methods, it shows that the proposed method has an average improvement of 8% and 8.9% in terms of superiority and stability compared with other algorithms. The optimized configuration scheme is determined by analyzing a practical application case, and the results verify the effectiveness of the proposed method, which has practical application value. References | Related Articles | Metrics

System Modeling & Optimization Algorithm

Scheduling Optimization of Fully Flexible Multi-cabin Vehicles for Garbage Classification YANG Chunxia, WANG Hao, CHEN Haibo, WANG Xiaojun 2024, 27 (2):  130-137.  doi: 10.3969/j.issn.1007-7375.220184 Abstract ( 956 )   HTML ( 24 )   PDF (942KB) ( 163 )   Save Garbage classification and transportation is an important link in the effective implementation of garbage classification policies. The existing classified transportation methods mainly focus on dedicated vehicles with single cabins, which easily leads to the waste of vehicle resources and congestion in garbage collection areas. To this end, this paper introduces multi-cabin vehicles that can transport various kinds of garbage independently at the same time. Based on the actual situation of garbage collection and transportation in China, we analyze the setting of cabins as well as the mode of collection and transportation, etc. On this basis, a fully flexible multi-cabin vehicle routing problem (full flexible-MCVRP) and its optimization model are proposed, and the genetic algorithm is used to solve it. An actual example shows that compared with the existing scheme of dedicated vehicles classification scheduling, the multi-cabin vehicle scheduling scheme proposed in this paper has a shorter total travel route and higher efficiency of collection and transportation. Furthermore, focusing on several typical urban living areas, the influence of garbage classification ratio on the allocation scheme of vehicle cabins is studied, which provides guidance for designing multi-cabin vehicles. References | Related Articles | Metrics

Route Optimization of Multimodal Transportation for Containers in the Yangtze River with Uncertain Freight Rates LI Jun, WEN Xiang, LIANG Xiaolei 2024, 27 (2):  138-146.  doi: 10.3969/j.issn.1007-7375.220245 Abstract ( 228 )   HTML ( 32 )   PDF (833KB) ( 133 )   Save In order to fully adapt to scenario of container multimodal transportation in the Yangtze River, on the basis of analyzing the characteristics of such scenario, the impact of uncertain freight rates and navigation conditions of different route segments are taken into account in this paper. A route optimization model for container multimodal transportation in the Yangtze River with uncertain conditions is built with the objective of minimizing the total cost including transportation turnover cost, carbon emission cost and time cost. Also, a Monte Carlo simulation and an improved wolf colony algorithm are designed to solve the model. Finally, an example is given to verify the effectiveness of the improved wolf colony algorithm. The analysis results of the example show that highway transportation can effectively control the time cost of transportation, while railway and waterway transportation have more advantages in transportation economy and greenness. Considering the navigation capacity and the maximum number of transfers of the Yangtze River waterway, the transportation task may choose a transportation scheme with higher cost to meet the requirements of the waterway and shippers. References | Related Articles | Metrics

Vehicle Routing with Multiple Trips and Multiple Time Windows WU Tingying, LU Jiaqi, XIA Yang 2024, 27 (2):  147-157.  doi: 10.3969/j.issn.1007-7375.220254 Abstract ( 290 )   HTML ( 28 )   PDF (576KB) ( 204 )   Save Focusing on the multi-trip transportation of terminal logistics and the diverse demand of customers for service time, a vehicle routing problem with multiple trips and multiple time windows is studied. A bi-objective mixed integer programming model is established to minimize the number of vehicles and the total transportation cost, while an improved adaptive large neighborhood search algorithm is designed to solve the problem. In this algorithm, a variety of efficient destroy and repair operators based on three levels of routes, travel distances and customer points are constructed to expand the search space of solutions. An adaptive strategy is used to select efficient search operators and a simulated annealing rule is introduced to avoid the solution from falling into local optimum and improve the search efficiency. By analyzing the experimental results of instances with various scales, the advantages of the improved adaptive large neighborhood search algorithm are verified, and the positive impact of the model considering multiple trips on the total transportation cost is analyzed. References | Related Articles | Metrics

Exploration of Cutting-edge Technology

A Study on Digital Derivation Mechanisms and Maturity of Digital Twins in Manufacturing Firms WANG Feng, GAI Yongjie, ZHANG Haitao 2024, 27 (2):  158-172.  doi: 10.3969/j.issn.1007-7375.230153 Abstract ( 1029 )   HTML ( 35 )   PDF (1420KB) ( 276 )   Save This study explores the basic management architecture (composition, connotation, components, functions, etc.) of digital twins regarding "production power, computational power, digital power" in manufacturing enterprises, and analyzes the integrated process framework and big data function model based on the endogenous 4.0 value chain of digital twins. The 4.0 value chain has a value transfer and iteration mechanisms, which empowers manufacturing enterprises to digitize their supply chains and expand the industrial chain derivation. Manufacturing enterprises generate big data through digital twins, where a digital industrial chain is formed (4.0 value chains, digitized supply chains, mobile value-added service MvaS chains, demand chains, and spatial platform chains). This study innovatively analyzes a digital growth path (digital twins →4.0 value chains → supply chains → digital industry chains → digital economy) through the digital twin derivation mechanism of manufacturing enterprises. Also, a five-level digital twin maturity model and an evaluation algorithm for manufacturing companies is comprehensively designed and condensed. Digital twins reshape the theory paradigm of digital management, and tamp the digital era foundation (smart manufacturing, digital factories, digital enterprises, digital industry chains, digital economy, etc.), having a far-reaching impact on the development of the digital economy, with important research and application value. References | Related Articles | Metrics