Takeshi Tateyama

Service process simulation for integrated service evaluation

The design of services has become an important issue discussed in various industries. For the design of attractive and sustainable services, it is important to satisfy the requirements of various stakeholders simultaneously. To evaluate the fulfillment of them in a quantitative manner, the process simulation of a service would be an effective approach. However, the method to develop a simulation model to evaluate a service with various evaluation criteria has not been proposed. In this study, the authors propose the evaluation framework and the service process model to describe the influence of a service to various stakeholders and suggest the simulation method based on it.

A teaching method using a self-organizing map for reinforcement learning

We described a new preteaching method for re-inforcement learning using a self-organizing map (SOM). The purpose is to increase the learning rate using a small amount of teaching data generated by a human expert. In our proposed method, the SOM is used to generate the initial teaching data for the reinforcement learning agent from a small amount of teaching data. The reinforcement learning function of the agent is initialized by using the teaching data generated by the SOM in order to increase the probability of selecting the optimal actions it estimates. Because the agent can get high rewards from the start of reinforcement learning, it is expected that the learning rate will increase. The results of a mobile robot simulation showed that the learning rate had increased even though the human expert had showed only a small amount of teaching data.

Service Design Methodology for Cooperative Services

In accordance with the maturation of industry, the major mean of value offering shifts from mass products to services. To develop successful services, the cooperation with different organizations and individuals is effective since innovative services often require various capabilities and resources which are difficult to obtain by one company. For the realization of such a cooperative service, a design methodology which can handle requirements of various stakeholders and a complex service structure is required. However, the general methodology for cooperative services has not been established. In this paper, the authors propose a service design methodology of cooperative services. This methodology includes a service evaluation framework to adjust the specification of a service which influences the multiple stakeholders, a computational simulation method for service evaluation and a service design process based on the proposed methods.© 2011 ASME

A Reinforcement Learning Algorithm for Continuous State Spaces using Multiple Fuzzy-ART Networks

This paper describes a new reinforcement learning system for unknown continuous state space environments. The purpose of our study is to divide the continuous state space to enable a reinforcement learning agent to perform a task well. Our method uses multiple fuzzy-ART (adaptive resonance theory) networks to divide a continuous state space. In our method, multiple reinforcement learning modules that use the fuzzy-ART networks as state recognizers learn concurrently, and the agent changes the state spaces for action selection from low resolution to high resolution in order to realize a good balance between the speed of the learning and its optimality. The results of the mobile robot simulation show the usefulness and efficiency of our learning system

Parallel reinforcement learning systems using exploration agents and dyna-Q algorithm

We propose a new strategy for parallel reinforcement learning; using this strategy, the optimal value function and policy can be constructed more quickly than by using traditional strategies. We define two types of agents: exploitation agents and exploration agents. The exploitation agents select actions mainly for the purpose of exploitation, and the exploration agents concentrate on exploration by using the extended k-certainty exploration method. These agents learn in the same environment in parallel, combine each value function periodically and execute Dyna-Q. The use of this strategy, make it possible to expect the construction of the optimal value function , and enables the exploration agents to quickly select the optimal actions. The experimental results of the mobile robot simulation showed the applicability of our method.

Service Flow Simulation Using Scene Transition Nets (STN) Including Satisfaction-Attribute Value Functions

A new academic field, “service engineering” has emerged; it involves active investigation for increasing the productivities of service industries. However, there are only a few effective tools for the simulation and evaluation of complex services that have been designed using concepts from the field of service engineering. To overcome this shortcoming, the authors propose a multi-agent service flow simulation method using scene transition nets (STNs) that is a very useful graphical modeling and simulation method for application to discrete-continuous hybrid systems. This method treats services as complicated multi-agent and hybrid systems similar to manufacturing systems. To obtain realistic values of customer satisfaction, the authors input satisfaction-attribute value (S-AV) functions, which are often used in service engineering in service flow models using STN. This method visualizes discrete flows of services and temporal changes in the values of customer satisfaction and various other variables of service models. The authors present the results of the simulation of an online DVD rental service involving multiple agents to show the effectiveness of the proposed method that uses an STN GUI simulator developed by them.Copyright

A Service Flow Simulation Method Using Multi-aspect Scene Transition Nets (STNs) Modeling

Recently, a new academic field, “service engineering,” has been very actively investigated. However, there are few effective methods and tools to simulate and evaluate services designed based on the concept of service engineering. The authors previously proposed a service flow simulation method that uses scene transition nets (STNs), which is a graphic modeling and simulation method for discrete-continuous hybrid systems. However, this method does not consider the manner in which complex service flows are modeled and simulated, including multiple layered, parallel, and interrupted structures, and it is difficult to construct STN models of such complex systems using the existing STN concepts. In this paper, the authors propose a new STN modeling method that uses “multi-aspect STN modeling” concepts to provide easy methods for modeling such complex services. The experimental results for the modeling and simulation of an elevator monitoring.. service demonstrate the effectiveness of the proposed method.

Service flow simulation using reinforcement learning models and scene transition nets

Recently, a new academic field, ldquoservice engineeringrdquo has been very actively investigated. However, there are few effective software tools to simulate and evaluate services designed based on the concept of service engineering. In the past, the authors proposed a service flow simulation method using scene transition nets(STN) which is a graphic modeling and simulation method for discrete-continuous hybrid system. However, this method cannot simulate complex service flows including customerspsila decision-making. Nowadays, ldquoneuro economicsrdquo and ldquoneuro marketingrdquo have gotten a lot of attention as new study fields to understand customerspsila behaviors from a viewpoint of brain science. In these studies, it turned out that mechanism of reinforcement learning concerns behavioral selections of customers. In this paper, the authors propose to develop decision-making processes models of customers and to simulate customerspsila behaviors and service flows by using reinforcement learning models and STN.

Web-Based Knowledge Database Construction Method for Supporting Design

In recent years, comprehensive problem solving by artefact designers has been required as demands related to artefacts become greater and more complicated. In relation to this background, we have proposed Universal Abduction Studio (UAS), a computer environment that synthetically supports a creative design. However, in general, it is difficult for a designer to acquire multiple domain knowledge because this knowledge is accumulated manually. This paper proposes a Web-based knowledge database construction method for supporting design using UAS.

Service evaluation method for managing uncertainty

A service is mainly produced by human capabilities and their interaction, and therefore, the process for the service production includes a lot of uncertainties caused by human factors. In order for service organizations to cope with these uncertainties, in this study, the concept of the modular architecture is applied to the service organization. Especially, this study proposes a method to determine teams consisting of human resources based on the concept of the modular service organization. The effectiveness of this method is demonstrated by the application to a hotel service.