Manufacturing intelligence and innovation for digi

Authors: ChenFu Chien Mitsuo Gen Yongjiang Shi ChiaYu Hsu

Publish Date: 2014/03/21

Volume: 25, Issue: 5, Pages: 845-847

PDF Link

Abstract

Global manufacturing networks and supply chains are facing ongoing dynamic changes in the evolutionary business ecosystems The introduction of new process technologies and the advances in manufacturing intelligence capabilities are having profound effects on the strategies decisions management and operations involved in manufacturing Chien et al 2007 Leachman et al 2007 While big data is accumulated due to the fully automation manufacturing facilities and logistics systems for business integration various solutions and techniques are developed to extract useful information and derive effective intelligence to address new challenges such as cycle time reduction Kuo et al 2011 Chien et al 2011 defect diagnosis Hsu and Chien 2007 Liu and Chien 2013 demand forecast Chien et al 2010 equipment management Chien et al 2008 Hsu et al 2012 Chien et al 2013a human capital Chen and Chien 2011 and bio informatics Lin and Chien 2009 In particular manufacturing innovation and manufacturing intelligence technologies are developed to empower manufacturing excellence via soft computing decision technologies and evolutionary algorithms to construct intelligent algorithms and solutions that can be embedded in various information systems for enterprise resources planning ERP advanced production system APS advanced process control and advanced equipment control APC/AEC manufacturing execution system MES engineering data analysis EDA and supply chain management SCM to enhance decision quality and management effectiveness Chien et al 2012 2013b Wu et al 2012a b By seamless integration of intelligence and decision technologies combined recent IT technology manufacturing intelligence have been completely become the trend of This special issue of the Journal of Intelligent Manufacturing JIM aims to address the critical needs for future factories logistics and global supply chain for Manufacturing Intelligence and Innovation for Digital Manufacturing and Operational ExcellenceThis special issue is developed to exchange the foresights and disseminate the research results on Manufacturing Intelligence and Innovation for Digital Manufacturing and Operational Excellence The special issue presents insightful comprehensive theoretical practical and realworld applications of advanced decision and intelligence technologies in manufacturing and logistics Original high quality papers of theoretical developments and empirical studies with scientific novelty and insights were invited from the scholars presented in the 2012 International Symposium on Semiconductor Manufacturing Intelligence ISMI2012 January 6–8 2012 at the Hsinchu Taiwan while other papers were directly submitted for review and publication All the submitted manuscripts were carefully blind peer reviewed by at least two referees in accordance with Journal of Intelligent Manufacturing procedure and finally fourteen papers were selected for this special issueThe selected papers in this special issue are clustered into four interrelated groups to address the theme of this special issue from various perspectives The first cluster addresses the important issues for manufacturing intelligence methodologies for multiobjective scheduling In particular “Multiobjective evolutionary algorithm for manufacturing scheduling problems stateoftheart survey” by Gen and Lin addresses the design of multiobjective evolutionary algorithms MOEAs to solve a variety of scheduling problems including job shop scheduling JSP flexible JSP automatic guided vehicle AGV dispatching in flexible manufacturing system FMS integrated process planning and scheduling “Cooperative estimation of distribution algorithm a novel approach for semiconductor final test scheduling problems” by Hao Wu Chien and Gen proposes a novel cooperative estimation of distribution algorithm CEDA that incorporates a cooperative coevolutionary paradigm to extend the model features and improve the evaluation leadtime of estimation of distribution algorithm EDA outperforming hybrid GAs for several semiconductor final test scheduling SFTS problems “Hybrid Sampling Strategybased Multiobjective evolutionary algorithm for process planning and scheduling problem” by Zhang Gen and Jo develops a hybrid sampling strategybased multiobjective evolutionary algorithm HSSMOEA to solve process planning and scheduling problemThe second cluster focuses on complex challenges in modern production facilities for operation effectiveness In particular “Determining the operatormachine assignment for machine interference problem and an empirical study in semiconductor test facility” by Chien Zheng and Lin develops a methodology to determine the optimal assignment relationships between the test machines and the operators for different product mixes to enhance productivity and system performance in which an empirical study has validates the proposed approach “Therbligbased energy demand modeling methodology of machining process to support intelligent manufacturing” by Jia Tang and Lv develops a mathematical model of energy demand of machining processes by linking the activity and energy demand units with machining state Two cases including energy demand of turning process and energy demand of three grooving processes have shown its practical viability to estimate energy demand for the energyoriented intelligent manufacturing “Hierarchical indices to detect the tool condition change with highdimensional equipment data for semiconductor manufacturing” by Yu Lin and Chien investigates the ways to early detect the changes of equipment conditions for efficient equipment maintenance so that domain engineers can identify potential root causes by drilling down the proposed hierarchical indices “Integrated data envelopment analysis and neural network model for forecasting performance of wafer fabrication operations” by Hsu constructs a model that integrates data envelopment analysis DEA and a backpropagation neural network BPNN for performance evaluation and forecast in which an empirical study was conducted for a semiconductor company for validationThe third cluster introduces novel approaches and applications for manufacturing intelligence in various domains In particular “An empirical study of designofexperiment data mining for yieldloss diagnosis for semiconductor manufacturing” by Chien Chang and Wang proposes a novel approach integrating DesignofExperiment DOE and data mining to effectively extract “experimental designs” from big manufacturing data for narrowing the possible root causes that have been implemented as a function of engineering data analysis platform in the case company “Tissue characterization of coronary plaque by kNN classifier with fractalbased features of IVUS RFsignal” by Uchino Koga Misawa and Suetake proposes a tissue characterization method for coronary plaques by using knearest neighbor kNN method and fractal analysisbased feature that are collected from radiofrequency RF signals measured by the intravascular ultrasound IVUS method “The construction of a hospital disease tracking and control system with diseases infection probability model” by Huang and Xie addresses the application of Radio Frequency Identification RFID to monitor the infected patients and detect potential routes of disease transmission in which a tree structure algorithm and an infection probability model were employed to trace the transmission routes discover the correlations between carriers and suspected cases and derive the infection probability “GMVN oriented SBOX knowledge expression and reasoning framework” by Li Xie and Tang proposes an intelligent integration framework of Semantic Bill of X SBOX by combining with the technologies of ontology Semantic Web Rule Language SWRL and Bill of X BOX to integrate the structures and operation rules into bill models and thus transform the strategy decisions management and operations involved in Global Manufacturing Virtual Network GMVN into semantic reasoning activitiesThe fourth cluster solves critical management problems involved in intelligent logistics and supply chains “A multiobjective genetic algorithm for yard crane scheduling problem with multiple work lines” by Liang Guo Gen and Jo develops a multiobjective 0–1 integer programming MO01IP model considering the minimum total completion time of all yard cranes and the maximization balanced distribution of the completion time at the same time to solve the yard scheduling problem “An adaptive genetic algorithm for the time dependent inventory routing problem” by Cho Lee Lee and Gen proposes an adaptive genetic algorithm AGA to the time dependent inventory routing problem TDIRP in which inventory control and time dependent vehicle routing decisions for a set of retailers are made simultaneously over a specific planning horizon “Optimization technique by genetic algorithms for international logistics” by Takeyasu and Kainosho addresses a sea and air transportation problem by multistep tournament selection with considering transportation cost and warehouse stock fees in the objective functionWhile this special issue has successfully addressed some of critical research needs for manufacturing intelligence and innovations for digital manufacturing and operational excellence we would like to express our sincere appreciations to all of the authors the anonymous reviewers and the EditorinChief Professor Andrew Kusiak and the Assistant Babu Krishnamoorthy of Journal of Intelligent Manufacturing for their invaluable contributions and great supports to this special issue

Keywords: