Intelligent Systems in Operations: Methods, Models and Applications in the Supply Chain

Intelligent Systems in Operations: Methods, Models and Applications in the Supply Chain

Barin Nag (Towson University, USA)
Indexed In: SCOPUS
Release Date: March, 2010|Copyright: © 2010 |Pages: 386
ISBN13: 9781615206056|ISBN10: 1615206051|EISBN13: 9781615206063|DOI: 10.4018/978-1-61520-605-6

Description

Artificial Intelligence and other Intelligent Systems heavily contribute to evolution and growth in the field of Operations Management.

Intelligent Systems in Operations: Models, Methods, and Applications introduces current and original research in intelligent systems and methodologies. This book intends to provide knowledge and insights on present and future AI applications in OM from current research-oriented thinking on AI-based systems in the benefit of OM tools and decisions in terms of theoretical and empirical models, methods and their comparisons, and actual and proposed applications. A must have for industry professionals seeking examples of AI applications in OM, researchers looking for material that can be extended into further research, and graduate students in the field.

Topics Covered

The many academic areas covered in this publication include, but are not limited to:

  • Collaborative Supply Chain Management
  • Computational Intelligence
  • Decision Support Systems
  • E-Procurement
  • Intelligent Decision Systems
  • Intelligent Simulation Systems
  • Mathematical Models for Optimization
  • Multi Agent Systems
  • Project Management
  • Supply Chain Event Management (SCEM)

Reviews and Testimonials

This book presents a variety of current and original research in intelligent systems and methodologies that is directly relevant and is application oriented.

– Barin Nag, College of Business & Economics of Towson University

Table of Contents and List of Contributors

Search this Book:
Reset

Preface

The development and study of models, methods, and applications, of intelligent systems in operations is an area of research that is in rapid growth with increasing numbers of quality contributions. These contributions serve to develop and advance the body of knowledge in the field by the addition of new and improved intelligent methods, and by the use of models based on artificial intelligence techniques. Most of this research tends to be practical rather than theoretical, and extends existing applications or develops new applications in business and industry with benefits for all in terms of cost, productivity, and quality. This book presents a variety of current and original research in intelligent systems and methodologies that is directly relevant and is application oriented.

One of the key elements of the 21st century business environment is that business and industrial functions and operations are global in nature. Management decisions are often made on a global basis with consideration of global decision factors. Even when an operational decision is made locally, the impact of the decision can be global. Further, concerns about operational decision problems and the methods used to solve them vary from one country to another. However, the impact of the decision and the methods used might be effective in a different environment somewhere else in the world. This book presents work from many researchers from a number of countries around the work that are actively engaged in research on intelligent systems in operations. Some of this work is drawn from their experiences in their home countries but have impacts and applications in other countries as well. The roster of authors includes those who are resident in one country and are concerned with the problems there, but are really from another country and have an understanding of problems that span multiple countries.

The importance of the topic begins with the fact that an operation is at the core of every business and industrial process. A business exists for the single purpose of transforming inputs into outputs by some process that adds value to the product for its customers. The sale of the product to the customers generates revenue, and a profit that results from the differential of the revenue and the costs of the inputs. Even a non-profit organization operates on the same value added principle minus the generated profits. The concept of profitability of the organization depends on the productivity of the organization, and productivity is often defined as the quantity of output per unit of input. Productivity that is larger and more effective in terms of customer satisfaction leads to lower cost structures vis-à-vis the revenue pattern, and thus to greater profitability.

Another factor in the effectiveness of the operational business transformation process is the supply chain and the management of the supply chain. The supply chain may be defined as the sequence of supplies of input components and transformation processes from the initial items that may be derived from the ground in mining operations or from farming operations, all the way to the final products delivered to the customer or end user, in a manufacturing or service context. Whether complex or simple, the importance of the supply chain is that the supply chain cost structure dominates the process sequence, and significant advantages may be obtained by managing supply chain strategies and procedures compared to forced increases in revenue (Heizer & Render, 2007). The supply chain issue is addressed by several papers in the book.

The major question raised by the productivity and supply chain concerns as raised above has to do with decision issues. It was stated by Herbert Simon in a paper that won him a Nobel Prize in 1956 that decision making is about problem solving. In other words, a decision in a business process needs to be made only when there is a problem, i.e. a situation exists where the present output does not match the expected or desired output. Simon described problem solving in four phases labeled Intelligence, Design, Choice, and Implementation. The phases describe respectively gathering information, developing alternative solutions, selecting the best solution, and implementing, often in an iterative way (Turban & Volonino, 2009) . Simon’s approach led to the description of problems in a spectrum ranging from structured and programmable to unstructured and solvable only by intelligent entities, with semi-structured problems in the middle. The goal of intelligent systems is to develop and implement automated and systemic solutions for the vast range of semi-structured problems requiring some level of intelligence in the solver, and preferably extending in the direction of unstructured problems.

Decision making in the business environment, as commonly performed by managers around the world, commonly involve people, equipment, targets, and uncertainties arising from market demands, the supply chain, performance characteristics, and the financial environment. It is quite common that the problem environment is less structured than one would want it to be. Conventional or programmable decision making is often inadequate as a computerized process. The field of operations management is one of the business decision areas having many problems designated as relatively unstructured because of uncertainties or other factors. The problems are important in the business context, and are in need of solution methodologies. Intelligent systems developed using concepts, tools, and methodologies, as used in Artificial Intelligence (AI) are transforming the world of Operations Management (OM) by the introduction of computerized and intelligent decision making tools, particularly in the areas of semi-structured decisions.

Intelligent systems are a collection of systems, methods, and technologies, derived from an established body of knowledge known as Artificial Intelligence, which is a way of replicating human intelligence and thinking in computer systems that are artificial or man-made. The applications of AI algorithms and intelligent systems embedded in various methods and devices are transforming operations around the world by increasing profitability and performance, by reducing the lead times for delivery, by making new manufactured and service products widely available at reduced costs and times to a larger number of customers, by improving performance, cost, and quality in the supply chain from the source to the customer, and by globalizing the supply systems and the customer base. This subject area is a bridge between the two traditional areas of Operations Management and Information Systems/Computer Science. The present focus is on the applications of Intelligent Systems in Operations Management (OM).This publication will present a variety of new research and new ideas in this context, as proposed systems, or as operational models, or as suggested and implemented methods in a variety of applications. Included are comparisons of modeling and application strategies in a global context. It is intended to cover all aspects of operations and the management of operations. Some of these functional areas are listed as Supply Chain Management, Manufacturing and Distribution with Logistics, Financial Services of all types, Healthcare and Medical Services, and all other manufacturing and service operations not mentioned above.

This book will provide knowledge and insights on present and future AI applications in OM from current research-oriented thinking on AI-based systems in the benefit of OM tools and decisions in terms of theoretical and empirical models, methods and their comparisons, and actual and proposed applications. It is intended to be inclusive with respect to all the diverse concerns of OM functions and issues, and will be global in its perspective. The diversity of content areas and applications is evident in the content. A brief look at the author biographies and affiliations shows the global perspective of the research presented. In summary, the book is not comprehensive in terms of AI methods in OM, as no book ever can be, but it presents a variety of research for thought in a variety of research and study areas, in a variety of applications, to support a variety of research.

The book will benefit professionals in industry seeking examples of applications, researchers looking for material that can be extended into further research that is as valuable as that found here, and graduate students in the field learning to do research and looking ideas and models as examples to build upon. The audience includes diverse application areas from industrial Operations Management to Business Process Management to Supply Chain Management, and also on the computational and algorithmic side from Computer Science to Information Science and Information Technology. In addition to covering all relevant functional areas as listed above, the book provides insights and executive support to executives in relevant industries tasked with decision making in these areas, and the management of expertise, methods, and knowledge, as it relates to these concepts.

Summary descriptions of the chapters are given below in alphabetical order with the first author. It is intentional that these are not in any sequence or progression to accommodate the various preferences of readers that have different outlooks and priorities:

Barnes, A.P. & Hammell, R.J. Employing Intelligent Decision Systems to Aid in Information Technology Project Status Decisions. The initiation of an information technology (IT) project by a firm, and the investment decision, usually has an objective of realizing benefits in the informational, strategic, transactional, and infrastructural objective areas of its portfolio. In the project management perspective, the important concerns are the performance of the project from the viewpoints of scope, schedule, cost, and other constraints. It is common that IT projects fail to meet objectives. This paper presents a case-based reasoning decision support architecture to provide a collaborative intelligent agent system to aid in recommendations of project status using color indicators derived from the progress and condition of the project-related constraints.

Camarao, C., Galvao, M., &Viera, N. SAT and Planning: An Overview This chapter studies the Satisfiability Problem (SAT), reviewing its importance in a wide range of applications. One application in Operations Management is planning. First, a review is presented of methods currently employed by modern SAT-solvers. Next, classical planning is used as an illustrative example to demonstrate the translation of a significant problem into SAT. Important results and studies are stated concerning reductions of planning into SAT. Finally, an explanation is given about the construction of a SAT instance which is satisfiable if and only if an instance an instance of a bounded version of the classic blocks-world problem is solvable.

Chen, W. & Decker, K.S. Integrated Multi-Agent Coordination. The key topics of planning and scheduling from Operations Research and Multi-Agent Systems have typically focused at an abstract system level on the development of interaction protocols to be imposed on agents. The internal task structures of individual agents that affect the higher-level coordination behaviors have been less studied. Collaborative multi-agent planning addresses problems such as uncertainties in plan outcomes, anticipation of likely contingencies, and evaluations of the way agent actions achieve worth-oriented goals. This paper presents extensions and restrictions, called extended hierarchical task networks, to the traditional plan and schedule representations that allow the formal definition of an integrated multi-agent coordination problem. The chapter discusses open issues in multi-agent coordination, and proposes a general solution towards successful distributed goal achievement by analyzing the task structures of participating agents.

Dunkel, J., Bruns, R., & Pawlowski, O. Complex Event Processing in Sensor-based Decision Support Systems. The chapter presents a reference architecture for sensor-based decision support systems, which enables the analysis and processing of complex event streams in real-time. The proposed architecture provides a conceptual basis for the development of flexible software frameworks that can be adapted to meet the needs of various applications. The architectural approach is based on semantically rich event models providing the different stages of the decision process. The approach is illustrated in the domain of road traffic management in high-capacity road networks. The need for this work arises from the high volume of continuously generated sensor events that decision support systems have to deal with. Conventional software architectures do not explicitly target the efficient processing of continuous event streams. The high volume of events, and the complex dependencies, implies that it is not possible to have a fixed or predefined process flow at the business level. Complex Event Processing (CEP) has been proposed recently as a general process model for event streams. CEP can provide mechanisms for computing high event volumes, but it does not define methodologies, models, and reference architectures that would make event processing mature software architecture.

Hammell II, R.J., Hoksbergen, J., Wood, J., & Christensen, M. Computational Intelligence for Information Technology Project Management This chapter addresses the management of Information Technology (IT) projects, an area that has become difficult and challenging because of the growth in numbers and complexity of these projects. The magnitude of the problem is shown by the fact that over two-thirds of the IT projects fail to meet their objectives. This work uses Computational Intelligence (CI) methodology, an area of research focused on the development of intelligent systems to help with complex problems. CI is a process of integrating techniques and methodologies to assist in problem domains vagueness, approximations, and uncertainties, pervade the information, data, and even the problem itself. Leveraging the power of CI methods on IT project management is a logical approach to improving IT project success rates. Examined in this work are the core CI technologies of fuzzy logic, neural networks, and genetic algorithms, vis-à-vis current and potential future applications to assist IT project managers.

Ibarra-Rojas, O.J., Rios-Solis, Y.A., & Chacon-Mondragon, O.L. Piece-mold-machine manufacturing planning This chapter is a study of a manufacturing process that puts together pieces produced on molds mounted on machines. To address the primary concern of minimizing the non-fulfilled demand, an integer linear programming formulation is used to integrate the most important features of the problem. The solution is used to determine the quantities of pieces to produce and the allocation of molds to machines. The objective function is designed to maximize production so that the non-fulfilled demand is minimized. The problem belongs to the NP-hard class and is not solvable in an exact way. A heuristic solution methodology is used based on an Iterated Local Search Algorithm. Computational experimentation is used determine the manufacturing planning decisions, and to make conclusions about the difficulty.

Nag, B.N., Qiao, H., & Yao, D. Intelligent Simulation System for Supply Chain Event Management (SCEM) This chapter presents and discusses an intelligent simulation system for Supply Chain Event Management for the purpose of designing and re-engineering the supply chain. Supply Chain Event Management (SCEM) is a subtopic of supply chain management designed to analyze the risks associated with uncertain supply chains, and re-engineer risky supply chains that are not suitable for the usual analysis of supply chains. The simulation framework is discussed in a series of layers, consisting of a component layer, a process layer, an intelligent execution layer, and an output layer for the presentation of results. The functional design of the layers is discussed on the basis of a simulation experiment on a case study that is illustrative of the process. In addition, implementation issues are reported and discussed on the basis of this real case study.

Pimentel, B.S., Mateus, G.R., & Almeida, F.A. Mathematical Models for Optimizing the Global Mining Supply Chain This chapter is a discussion on the application of intelligent decision support systems in the management of operations in the mining industry. An integrated perspective on the supply chain approach to mining operations takes into account mining, railway, and port operations, as well as the logistics channels that support domestic and international supply chain operations. The operations involve production and distribution planning, and scheduling problems, which have been addressed individually. Mathematical programming is the basis for this discussion. The primary topics discussed here are: recent developments in the use of operations research in the mining industry, integrated approaches towards the development of decision support systems to support global mining supply chains, and possible solution approaches to integrated supply chain problems. The focus is on mathematical programming formulations supported and solved by heuristic simulation and artificial intelligence solution techniques.

Renna, P. Negotiation policies for e-procurement by Multi Agent Systems This chapter is about a special e-Business application known as e-Procurement, i.e. electronic procurement in the supply chain as commonly found in an e-marketplace approach in business-to-business (B2B) applications. B2B applications of e-marketplace follow the development of Information and Communication Technologies (ICT) that allowed enterprises to adopt the e-marketplace approach. These applications are providing added value to manufacturing operations in terms of increasing global performance. The e-marketplace implementation is still not easy because of the need of human participation in all stages of the B2B process. The value added services proposed in this chapter are: workflow design, Multi Agent System, and a negotiation approach. The negotiation approach is based on the auction environment, with simulations to test the proposed approaches. Results are presented for the simulations conducted in several scenarios to determine the best approach.

Sgarbi, M., Colla, V., & Bioli, G. A 3D vision-based solution for product picking in industrial applications This chapter is about computer vision applications in robotic systems in industrial applications. Computer vision has become a key application factor in manufacturing processes, affecting quality control, assembly verification, and component tracking. Robotic systems play an important role in the automation of production lines. The common applications of 2D vision systems for placing objects on conveyor belts, where robot manipulators grasp the objects using only 2D information, is being replaced by 3D applications which enable valid solutions in bin-packing applications where objects are randomly placed inside a box. There are many real-world applications where the 3rd dimension is a requirement for successful operation. An example is where the objects differ in height, or are placed in front of a camera manually without any constraint on the distance between the object and the camera. The solution possibilities presented by a 3D system are balanced by the higher costs and complexities compared to 2D vision systems. This work presents a monocular system useful for picking applications by estimating the 3D position of a single marker attached to the target object in the assumption that the orientation of the object is approximately known.

Wautelet, Y., Achbany, Y., Lange, J., & Tran, V. Developing a Collaborative Supply Chain Management Platform: a Service-Oriented Approach This chapter develops the concept of service-oriented computing and Service Oriented Architecture (SOA) that is becoming increasingly popular. It enables flexible designs and adaptable software systems that can be easily adapted to the demands of software customers. Supply Chain Management is a primary beneficiary of this concept. This paper is a proposal to apply ProDAOSS, a process for developing adaptable an open service systems. The application is directed towards an industrial case study in outbound logistics. ProDAOSS is conceived as a broader development methodology to cover the entire software development life cycle, and is intended as a plug-in for I-Tropos. Flexible business processes are typically generically modeled with different complimentary views at an analytical level. At first, an aggregate services view of the entire application package is offered. Then the services are split using an agent ontology, using the i* framework to represent it as an organization of agents. A dynamic view completes the documentation by offering the service realization paths. At the design stage, the service center architecture proposes a reference architectural pattern for services realization in ana adaptable and open manner. The work finally presents the implemented platform for a particular service to manage transport, so that a reader can realize how the developments have been achieved.

I would like to end this preface with a grateful note of thanks to every author who has contributed valuable research work, making the book that much more interesting and informative. State of the art research is always uncertain and risky. A research proposal, or a research plan, or even a commitment towards a research contribution, does not always result in one. One has to appreciate the work, and the commitment to work, of all the authors who have made contributions, often in a relatively short time. Each one was peer-reviewed. Some needed adjustments. In the end, all came out with flying colors.

At this time, I would also like to thank the reviewers, many of whom came from the list of authors. However, no reviewer had any idea of who the authors were. That is the honor code of reviewing, and the reviewers did a wonderful job, often returning the review in a very short time. I serve on the editorial boards of some journals and I really appreciate these fast reviews. I appreciate the time taken by the reviewers to do this.

One last comment refers to the global nature of the authors’ affiliations, reflecting the global nature of the interests and research works represented in this book. Globalization of operations is the nature of the world today, and is of importance to me in terms of research interests. Globalization is reflected most effectively in this book with contributions from many countries in several continents.

Author(s)/Editor(s) Biography

Barin Nag is a Professor in the College of Business & Economics of Towson University. He has a PhD in Management Science and Computer Science from the University of Maryland at College Park. He has Bachelor’s and Master’s degree in Electrical Engineering from the University of Calcutta, India. He has over forty publications in artificial intelligence applications in logistics, production scheduling, financial management, and supply chain management.

Indices