Browsing by Author "Kaihara, Toshiya"

Now showing 1 - 11 of 11
  • Item
    Agent-based adaptive production scheduling - a study of cooperative-competition in federated agent architecture
    (Springer-Verlag, Tokyo, 2009) Opadiji, Jayeola; Kaihara, Toshiya
    An increasingly popular method of improving the performance of complex systems operating in dynamic environments involves modeling such systems as social networks made up of a community of agents working together based on some basic principles of social interaction. However, this paradigm is not without its challenges brought about by the need for autonomy of agents in the system. While some problems can be solved by making the interaction protocol either strictly competitive or strictly cooperative, some other models require the system to incorporate both interaction schemes for improved performance. In this paper, we study how the seemingly contradictory effects of these two behaviours can be exploited for distributed problem solving by considering a flexible job shop scheduling problem in a dynamic order environment. The system is modeled using federated agent architecture. We implement a simple auction mechanism at each processing center and a global reinforcement learning mechanism to minimize cost contents in the system. Results of simulations using the cooperative-competition approach and the strictly competitive model are presented. Simulation results show that there were improvements in cost objectives of the system when the various processing centers cooperated through the learning mechanism, which also provides for adaptation of the system to a stream of random orders.
  • Item
    Distributed production scheduling using federated agent architecture
    (Springer-Verlag, Berlin,, 2007) Opadiji, Jayeola; Kaihara, Toshiya
    Making a production system readily reconfigurable in a bid to adapt to very fluid demand profile is pertinent to cost reduction and facility utilization objectives of the system. We consider a production scheduling methodology based on federated agent architecture designed for a flexible job shop with dynamic demand. The interaction protocol within the social network is based on a facilitated auction mechanism. Facilitator agents are responsible for coordinating information flow between a controller agent and processors in different job centers. These facilitators interact in such a way as to reduce tardiness of tasks once an order is accepted by the controller agent. This model employs competition at job centers to maximize financial returns and uses cooperation among facilitator agents to minimize weighted tardiness.
  • Item
    Market-based strategic procurement planning in convergent supply networks
    (Institute of Systems, Control and Information Engineers, Japan., 2009) Opadiji, Jayeola; Kaihara, Toshiya
    Convergent manufacturing supply networks facilitate the flow of complementary resources which make the application of competitive equilibrium algorithms difficult. In this work, we present a market-based decentralized approach which uses a market-oriented programming algorithm to obtain Pareto-optimal allocation of resources traded among agents which represent enterprise units in a supply network. It is assumed that every enterprise have unique preferences for suppliers in the markets in which they trade. Description of enterprise agents and the interaction protocol used to generate procurement plans for the enterprise units are presented.
  • Item
    A model for capturing the effects of Macroeconomic indicators on aggregate planning in a supply network
    (Tokyo Institute of Technology, Japan., 2006) Opadiji, Jayeola; Kaihara, Toshiya
    As competition in the global market is getting tougher, firms are turning to their supply networks in order to improve their competitiveness. This is done by improving the performance of some components of the supply network while not violating the requirements of the other components. This is undoubtedly a difficult task to achieve considering the autonomous nature of these units of the network and also their distributed topology. Within the last few years, firms have resulted into a globalization of their industrial workspace in order to put a lid on cost as well as harness the opportunities of new markets. The globalization of these workspaces however introduces new challenges to the supply chain of these firms in the form of impediments to trade across national borders. We present a model for estimating the effects of some of these macroeconomic indicators on a firms supply network. We define the supply network as a virtual enterprise network which has the ability of taking advantage of the technology provided by existing telecommunication networks including the internet, to conduct transactions and also provide for easy reconfiguration across enterprise boundaries. In this model, enterprise units are represented as trading agents operating within a competitive market structure to arrive at a paretoallocation of resources in the network. We first describe a process of trading among the agents within a Walrasian market structure and then we consider the effects of two macroeconomic variables – per-capital income and interest rates – on the pareto-allocation of resources in the virtual enterprise network.
  • Item
    Optimal Resource Allocation in Supply Network with Competitive Market Concept
    (CIRP International Seminar on Manufacturing Systems, University of Liverpool, UK, 2007) Opadiji, Jayeola; Kaihara, Toshiya
    We propose a competitive market-based model which uses price mechanism in an economy to determine an optimal allocation of resources among a set of trading agents. A supply network is represented by a set of production agents, which make use of the resources available to them in the market to produce an output defined by their technologies and demand agents seeking to maximize returns based on manufacturing budget. We take into consideration, changes in quantity of primary production factors – capital and labour – at facility locations. The resource allocation table obtained from simulation is a measure of value of market resources and facility capacities in the supply network.
  • Item
    A Proposal on agent-based production planning in integrated supply network
    (Elsevier Ltd., 2008) Opadiji, Jayeola; Kaihara, Toshiya
    We propose models that focus on the improvement of flexibility in manufacturing supply networks by enabling a tighter information coupling between the various planning levels without tampering with the autonomy of enterprises which are geographically distributed. The problem is approached from the perspective of social network planning using a community of agents. These agents have unique properties which they exhibit at different planning levels. Characterization of agents in the models is discussed.
  • Item
    A study on resource allocation optimization in a multi-commodity virtual enterprise network
    (Published by the Graduate School of Science and Technology, Kobe University, Japan., 2007) Opadiji, Jayeola; Kaihara, Toshiya
    Finding an optimal solution to various constrained optimization problems in the supply network of organizations is clearly a Herculean task considering the fact that some of the objectives to be optimized have conflicting requirements. A popular method used in tackling issues aimed at improving the performance of a supply network is to focus on a particular planning horizon and explore activities where improvements can be made. Optimization problems within a supply network vary from strategic to operational domains and a number of models have been developed to address these issues including probabilistic models, mathematical algorithms and some meta-heuristic methodologies. We propose a strategic decision support model based on Multi-Agent System (MAS) paradigm to find a pareto-optimal resource allocation among a group of trading agents in a multi-commodity market scenario. We employ the price mechanism model in microeconomics to create a trading floor for a set of reactive agents to search for a solution to a resource allocation problem in a virtual enterprise network. While the concept of Competitive General Equilibrium (CGE) is not new in the neo-classical economic domain, its application to solving resource allocation problem within a supply chain structure may not be as common. We first study some existing algorithms for CGE based on the Walrasian market model and then introduce an adaptation of the Market-Oriented Programming (MOP) paradigm to the supply network problem. We compare the performance of these algorithms and discuss application areas of the competitive market based approach.
  • Item
    A STUDY ON THE EFFECTS OF CONSUMER PREFERENCE ON PROFITABILITY IN A MASS CUSTOMIZATION SUPPLY CHAIN MODEL
    (Japan Joint Automation Control Conference, Kobe University, Japan, 2006) Opadiji, Jayeola; Kaihara, Toshiya
    In a bid to improve their competitiveness, there is a rapid growth in the number of firms adopting the “customer-is-king” philosophy by embedding mass-customization into their goods and services delivery network. The adaptation of essentially generic products to suit customer needs results in goods passing through the same production line having varying prices depending on the specificity of the products. In this paper, we discuss a model for studying the effects of consumer behaviour as indicated by their preference indices on the available products and variation in their purchasing power, on the profitability of a firm. This market-based framework gives a monetary-based insight into some issues like innovation, competitive advantage window, customer disposition to premium prices, etc, which planners consider during mass-customization of products. Results from simulations based on Market-Oriented Programming are presented.
  • Item
    Supplier selection in virtual enterprise model of manufacturing supply network
    (Springer-Verlag, Berlin, 2009) Kaihara, Toshiya; Opadiji, Jayeola
    The market-based approach to manufacturing supply network planning focuses on the competitive attitudes of various enterprises in the network to generate plans that seek to maximize the throughput of the network. It is this competitive behaviour of the member units that we explore in proposing a solution model for a supplier selection problem in convergent manufacturing supply networks. We present a formulation of autonomous units of the network as trading agents in a virtual enterprise network interacting to deliver value to market consumers and discuss the effect of internal and external trading parameters on the selection of suppliers by enterprise units.
  • Item
    Value delivery in convergent supply network with mediating auctioneers
    (Society of Instrument and Control Engineers, Japan., 2008) Opadiji, Jayeola; Kaihara, Toshiya
    Value delivery is the ultimate objective of every supply network. In this study, we propose a method of mitigating the effects of excessive competition in a convergent manufacturing supply network. Using an auction based protocol; market equilibrium is reached for a resource allocation problem within the network. In order to improve the quality of the solution obtained, a mediation method facilitated by network auctioneers is proposed. This method will allow an auctioneer to redistribute cost in its market in such a way as to improve throughput in that market which ultimately improves the throughput of the supply network.
  • Item
    Virtual enterprise model for integrated planning in manufacturing supply network
    (Springer-Verlag, London,, 2008) Opadiji, Jayeola; Kaihara, Toshiya
    Value creation in manufacturing supply network often involves complex interactions among autonomous enterprises. Consistency is required in decisions made in all the planning stages if enterprises in such a supply network are to continuously have feasible and flexible plans which respond to volatile trading environments. We propose a virtual enterprise model based that integrates the strategic and tactical planning stages of materials procurement activity in a manufacturing supply network. In this model, each enterprise is represented as an agent participating in economic activities in multiple virtual markets. We employ a market-based approach to find satisficing tactical plan for in a supply network given strategic constraints.