Comparative evaluation of genetic algorithms for job-shop scheduling

Many optimization problems from the industrial engineering world, in particular the manufacturing systems, are very complex in nature and quite hard to solve by conventional optimization techniques. There has been increasing interest in imitating living beings to solve such kinds of hard optimization problems. Simulating the natural evolutionary process of human beings results in stochastic optimization tech niques called evolutionary algorithms, which can often outperform conventional optimization methods when applied to difficult real-world problems. There are currently three main avenues of this research: genetic algorithms (GAs), evolutionary programming (EP) and evolution strategies (ESs). Among them, genetic algorithms are perhaps the most widely known types of evolutionary algorithms today. During the past years, several GAs for the job-shop scheduling problems have been proposed, each with different chromosome representation. In this paper, the different GAs are collected from the literature and an attempt has been made to evaluate them. The benchmark problems available in open literature are used for evaluation and the performance measure considered is makespan. The algorithms are coded in C+ +.     

A simple two-agent system for multi-objective flexible job-shop scheduling

Journal of Combinatorial Optimization - In this paper, a simple two-agent multi-objective scheduling system for flexible job-shop scheduling problem is proposed and corresponding framework is...     

Economic view of CIM system architecture

The synergistic advantage of systems integration for Computer Integrated Manufacturing (CIM) has been a powerful thrust behind the productivity advancement in recent decades. However, the dependency among components, the intangible benefits and complexity of the system have been a barrier for system developers and the business community. This paper reports a systematic approach to assess economic merits of the CIM system. It projects CIM system designs on a multiple layer architecture with consideration of project life cycle. It introduces a modelling formalism to present a structured view of the economic aspect of the CIM system. The multiple criteria's aspect of CIM justification was treated with the AHP (Analytical Hierarchy Process) method. The complexity of various cost factors is reduced through application of the ABC (Activity Based Costing) method. The benefits of low level activities, both tangible and intangible, are then hierarchically aggregated into higher level system objectives. It also provides the necessary linkages between the economic and other views. A simple example was used to illustrate the approach.     

Two issues in the evaluation of job-shop schedules

Two issues are examined concerning the evaluation of job-shop schedules. First, machine and labor cost/utilization are suggested as alternatives to physical utilization criteria. Secondly, an exploration is made of the use of Kiviat Charts as a means of integrating multidimensional criteria for the overall evaluation of the quality of job shop schedules.     

KBSS: a knowledge-based job-shop scheduling system*

Abstract. A knowledge based job-shop scheduling system (KBSS), which employs artificial intelligence techniques to solve general job-shop problems, is presented. The system consists of two main components: a knowledge base (KBSS model) and a control mechanism. The KBSS model has a three-level architecture which captures, in frames, knowledge about the job-shop domain and the characteristics of individual job-shop problems. The control mechanism consists of two modules: the primary scheduler (PS) and the heuristic scheduler (HS). PS is a rule-based scheduler built around a time-transcending heuristic which attempts to minimize the makespan and meet the due dates. If PS fails to meet the due dates then HS applies a combination of A* and beam search techniques in the space of alternative solutions and returns the best alternative it finds. The emphasis on due dates is a distinctive feature of KBSS. Experience with evaluating the presented scheduling system is reported and discussed.     

Production planning and control within a CIM environment: some current developments and requirements for the future

Abstract. This paper considers the integration problems associated with incorporating production planning and control (PP&C) systems within CIM; these problems are typically not differentiated from functionally related problems, and have in the past received relatively little attention. The on-going emergence of a number of standards relevant to manufacturing software systems provides an opportunity to assess the current situation with regard to proprietary PP&C systems and to suggest ways forward towards ‘seamless’ integration of PP&C within the CIM environment. The problems arising from conventional integration approaches are briefly described, and the concept of a simplifying ‘integration platform’ is introduced; the three-schema information architecture is presented as part of this ‘integration simplification’ approach. Current and developing PP&C systems are assessed against a number of information system integration criteria, and the results of a small survey of proprietary packages are presented. The paper concludes that information must be regarded as an enterprise-wide asset, rather than belonging to individual software applications, and that the adoption of relevant standards will enable seamless integration within a three-schema architecture. An increasing number of proprietary packages are adopting standards and methodologies which ease the integration problem significantly.     

A framework for production rescheduling in sociotechnical manufacturing environments

No plan survives contact with reality. Despite the rich research base regarding handling uncertainty in production planning and control systems, there is an intellectual gap between theory and practice with regard to handling unforeseen events generated by internal and external factors, such as unforeseen machine downtimes and changes in demand. Motivated by longitudinal observations in two industrial settings and an analysis of the relevant literature, a framework for rescheduling decision-making in the face of unforeseen production events is proposed. In practical settings, the effectiveness of decisions depends on a set of situational factors. The findings of this research can be utilised further to provide guidelines for developing effective decision support principles and systems, addressing the needs of organisational decision-makers.     

Priority class scheduling: Production scheduling for multi-objective environments

A new scheduling framework (priority class scheduling) is devised which reduces the conflict in the manufacturingmarketing interface. Priority class scheduling rewards manufacturing for making small improvements in efficiency, involves marketing in the setting of production priorities, improves customer service by putting in place realistic production timetables, allows manufacturing to concentrate on improving both quality and efficiency, and improves both the communication and cooperation between the shop floor, the production scheduler, marketing, and customer service. Within a production period, all jobs are classified into priority classes based on both their due dates and importance. Production is scheduled within each production period to maximize throughput subject to the priority class constraints- if any job in priority class i is started in the production period, then all jobs in priority class i - 1 must be completed within the production periods.     

A new value stream mapping approach for healthcare environments

Hospitals worldwide are giving a growing emphasis to the application of lean concepts in the healthcare sector, commonly known as ‘lean healthcare’. A fundamental tool that allows such implementations is the value stream mapping (VSM). The problem is that VSM models used in implementations of lean healthcare are simple adaptations of the original VSM model, which was initially directed towards manufacturing and may not always represent important support activities for the patient flow that directly impact treatment time. Within this context, this paper presents a new VSM approach for healthcare environments. This new VSM model, specifically designed for healthcare environments, contemplates all activities that directly affect the treatment time. In addition, the present paper also presents an action research in a Brazilian hospital where the proposed VSM model is compared to other VSM models found in the literature. The results shown that the proposed VSM model was able to identify some operational bottlenecks and wastes that interfere in the patient’s treatment that could not be identified by other mapping models studied.     

Requirements for successful implementation of new manufacturing technologies

This paper addresses the issue of why many U.S. companies have had major problems implementing “programmable automation” technologies in a manner to exploit their full potential. Many competitors in Japan and Europe have been much more successful. The theme of the paper is that the real impediment to the effective use of this new manufacturing hardware lies in some deeply entrenched attitudes and ways of doing things that are incompatible with the requirements and unique capabilities of the new hardware. Barriers to successful implementation are explored and implications for managerial practice are discussed.     

Guidelines for Hoshin Kanri implementation: development and discussion

Hoshin Kanri is a strategic management framework originally conceived in Japan and recognised for developing a deployment process that integrates business strategy and operations execution. Recently, there has been a resurgence of interest regarding Hoshin Kanri implementation among western managers, who usually face cultural challenges towards its use. This leads to the necessity of systematising a set of universally applicable guiding principles or guidelines, for Hoshin Kanri initiatives. This paper presents the results of an in-depth assessment of Hoshin Kanri guidelines carried out through a systematic process of experts? interviews. It uses as its starting point a model of the guidelines that was previously developed through a systematic literature review. The revising and refining of the guidelines is carried out by discussing their conceptual foundations and 20 guidelines are presented that may aid in the implementation of Hoshin Kanri in dfferent types of organisations.     

A multistage approach for production and workforce planning in long-cycle product environments

Aggregate production planning (APP) has been studied extensively for the past two decades. The APP problem, also called production and workforce scheduling, is to determine the optimal workforce and production level in each period of the planning horizon in order to satisfy demand forecasts for these periods. The advantages of the APP are low cost of data collection and computational cost of the running model; the accuracy of data; and, effective managerial understanding of the results. If the product of concern takes longer than one period, it is called a long-cycle product. Examples of long-cycle products are aircraft, ships, buildings and special machines. A detailed model incorporating dynamic productivity and long-cycle products considerations is presented to solve the problem of production and workforce planning. Using a multistage production system approach, a search technique is developed to solve this class of problems where the objective function is linear and some of the constraint coefficients are dynamically nonlinear. The model provides a better solution than an aggregate production planning model, often used to solve these problems.     

Beyond products: new strategic imperatives for developing competencies in dynamic environments

Strategy research has often taken a product-centered perspective. When firms compete in environments characterized by accelerating product life cycles, mass customization, and technological discontinuities, a product-centered perspective on strategy may help explain a firm’s current competitive advantage. However, this perspective adds little guidance in making strategies that create competitive advantage in the future. In this paper, we present a perspective in which dynamic environments require firms to focus on (1) building market-driven, technological, and integration competencies, not a stream of product improvements, and (2) decoupling these competencies from current products in order to create and exploit new opportunities. We discuss the perspective of products as a temporary integration of market and technology trajectories. Research propositions are presented and future implications are discussed.     

Effective eProcurement implementation process

As eProcurement is gaining popularity in business practice, researchers as well as practitioners have addressed the need of defining a proper eProcurement implementation strategy. This article aims to build an eProcurement implementation framework, which considers the main variables that have an impact on the eProcurement success and the links among them. The goal is to identify the main decisions that companies should take during the three principal phases of the eProcurement adoption (pre-implementation; implementation and post-implementation). Six case studies at different levels of eProcurement adoption have been used to identify the peculiarities of each phase of the process and to provide contingent guidelines.     

MRP system performance under lumpy demand environments

Material requirements planning (MRP) systems are deemed to deal with master schedules with lumpy demand patterns better than any other production scheduling system. Past studies have advocated important advantages of using MRP systems. The objective of this paper is to look into the impact of patterns of demand lumpiness on the performance of MRP systems by a simulation study. Results show that there is an important threshold point in terms of degree of lumpiness at which MRP system performance starts to deteriorate in the operating conditions considered. If master production schedules (MPS) can be controlled by manufacturers, MRP users should exercise caution to introduce demand lumpiness in MPS to improve system performance. If not, MRP users should then examine the given lumpiness and choose an appropriate lot-sizing rule that has been shown to take advantage of the effect of demand lumpiness.     

网络环境下平台企业的运营策略研究

平台企业在网络环境下发生了深刻的变革,选取恰当的视角研究它的运营策略是个重要的科学问题.以平台生态圈的实际规模、期望规模及时间为参数,构建平台生态圈生命周期曲线,通过识别关键节点将曲线分为若干发展阶段,平台生态圈的规模、网络效应、发展模式在各阶段中体现出不同的特征,这些特征为平台企业制定运营策略提供了依据.总体上讲,当平台企业无法充分利用网络效应时,应依托各种“非网络效应诱因”吸引用户进入生态圈,不断激发网络效应;反之,则应充分利用网络效应,迅速扩张规模,同时解决成长中暴露出的问题,优化成长方式.构建了平台企业运营策略的理论框架,深化了平台生态圈生命周期理论,对网络环境下的平台企业进行自我诊断、制定运营策略具有指导意义.     

Business games and implementation research

This paper reviews some recent research in OR implementation and suggests that a new approach is required, which will take fuller account of the perceptions of managers concerned with the problem. Research games in business settings seem to offer a means for such approach. Some lessons learned from a business game used in this way are described and their implications for modelbuilders are discussed.