Evolutionary Computing and Artificial Intelligence

Inhaltsverzeichnis

1. Frontmatter

2. This Is How I Feel About Complex Systems

   Takao Terano

   Abstract

   It has been around for nearly 40 years since I joined with the academic world in computer science related areas. Because I prefer to concrete real world topics, I, as a system scientist, usually research and develop challenging cutting-edge applications on complex systems. By complex systems, I do not only mean such areas as complex adaptive systems, chaotic, nor fuzzy systems, but, seemingly complicated systems required to be socially implemented. In this short paper, based on my final lecture at Tokyo Institute of Technology, I would like to present how I feel about complex systems and discuss principles of agent-based modeling, interdisciplinary research, and system creation. The paper concludes some comments on what we must do in the future.

3. Artificial Intelligence Technology and Social Problem Solving

   Yeunbae Kim, Jaehyuk Cha

   Abstract

   Modern societal issues occur in a broad spectrum with very high levels of complexity and challenges, many of which are becoming increasingly difficult to address without the aid of cutting-edge technology. To alleviate these social problems, the Korean government recently announced the implementation of mega-projects to solve low employment, population aging, low birth rate and social safety net problems by utilizing AI and ICBM (IoT, Cloud Computing, Big Data, Mobile) technologies. In this letter, we will present the views on how AI and ICT technologies can be applied to ease or solve social problems by sharing examples of research results from studies of social anxiety, environmental noise, mobility of the disabled, and problems in social safety. We will also describe how all these technologies, big data, methodologies and knowledge can be combined onto an open social informatics platform.

4. A Formal Model of Managerial Decision Making for Business Case Description

   Masaaki Kunigami, Takamasa Kikuchi, Takao Terano

   Abstract

   This paper proposes a novel formal description model of organizational decision-making: the Managerial Decision-Making Description Model (MDDM). This model introduces a four terminal element representation to describe managerial decisions redefining relationship between their objectives and resources. It enables them to compare various decision-making processes from not only actual business cases but also virtual ones, from an agent-based simulation, too. This model is also applicable in facilitation support for business gaming.

5. Evolutionary Computation Meets Multiagent Systems for Better Solving Optimization Problems

   Vinicius Renan de Carvalho, Jaime Simão Sichman

   Abstract

   In this work, we discuss the synergy between Evolutionary Computation (EC) and Multi-Agent Systems (MAS) when both are used together to enhance the process of solving optimization problems. Evolutionary algorithms are inspired by nature and follow Darwin theory of the fittest. They are usually applied where there is no specific algorithm which can solve optimization problems in a reasonable time. Multi-Agent Systems, in their turn, are collections of autonomous entities, named agents, that sense their environment and execute some actions in the environment to meet their individual or common goals. When these two techniques are applied together, one can create powerful approaches to better solve optimization problems. This paper presents an overview of this combined approach, considering both mono-objective and multi-objective approaches. In particular, we stress the importance of hyper-heuristic approaches, i.e., heuristics that help to choose the best EC algorithm among a candidate set.

6. A Practice Report on the Active Learning Using Business Game for the Teacher Training Students

   Hikaru Uchida, Katsutoshi Yuasa

   Abstract

   This paper describes a practice of business game using ICT (Information and Communication Technology, hereafter in ICT) prepared for the teacher-training course student. Since the Japanese primary and the secondary education, the environment of ICT in the classroom is improving. Furthermore, the forthcoming educational guidelines from the ministry are insisting on the active learning with ICT; it is more important to learn the active learning using ICT than ever. From the questionnaire survey, we examined how the students can use ICT for learning and what kind of difficulties the students have. As a result, it was speculated that students feel anxiety and difficulty in becoming teachers who do such classes because students have no experience of receiving active learning using ICT. In this study, as an example of active learning using ICT, we aim to make students experience business games using computer agents and aim to think more deeply about the possibility of using ICT for learning. In this paper, we describe the possibility of learning effect given by the design of the game using the computer agents. And we report the practice of business game using strategy agent. In the future, some concrete methods with ICT are also required such as mounting facilitating agents and simulating player agents in the game. This is research in progress.

7. Framework of Evaluating Business Partner Recommendation Beyond Industry Types Toward Virtual Corporation

   Taisei Mukai

   Abstract

   We propose a framework to evaluate a recommendation of unknown partners in an inter-business market by an artificial intelligence (AI) and a simulation. The reason is that unknown partner recommendation by AI such as a data mining or machine learning is difficult to evaluate because it is not possible to know correct partners in the real world. The framework is a flow of (1) proposing a method of recommending unknown business partners, (2) installing the recommendation method as AI into a firm agent, and (3) evaluating a recommended business partner by comparing performance between a recommendation method (machine learning, etc.) and an agent-based modeling simulation (ABMs). Since this framework can handle experiments assuming future situations, managers in a firm are possible to consider and judge recommendation methods and recommended partners according to virtual market conditions.

8. Analysis of Researchers Using Network Centralities of Co-authorship from the Academic Literature Database

   Masanori Fujita

   Abstract

   Finding and encouraging young promising researchers is crucial to develop science and technology and to promote innovation. In this paper, I am to clarify requirements for researchers to conduct organizational Research and Development (R&D) and propose a quantitative method to evaluate researchers that satisfies the requirements to evaluate researchers in organizational R&D fields. A questionnaire survey was conducted to R&D institutions in life science and information technology fields to clarify the required competencies and careers of researchers for organizational R&D projects. The result of the survey suggests that the institution members require the researchers’ competencies on not only “expertise of the research fields” but also “cooperativeness with others in the projects”. Based on the result, I focus on network centralities of co-author networks and propose a new quantitative method to evaluate researchers by measuring the network centralities from the academic literature database.

9. Debriefing Framework for Business Games Using Simulation Analysis

   Takamasa Kikuchi, Yuji Tanaka, Masaaki Kunigami, Takashi Yamada, Hiroshi Takahashi, Takao Terano

   Abstract

   Researchers are aware of the importance of debriefing in gaming. However, there has been little progress in establishing a methodology for evaluating player behavior. Therefore, we propose a framework to support the evaluation of player behavior in business games using computers. Specifically, we introduce a simulation analysis methodology that involves the following steps: (1) constructing an agent-based model based on the subject of the business game and categorizing simulation-logs; and (2) mapping logs of player behavior onto typed results. In this way, the positioning of the player in the overall simulation scenario is visualized, and a range of possible results is presented. Both players and facilitators receive information that is useful for debriefing.

10. Applications of Evolutionary Computation and Artificial Intelligence in Metallurgical Industry

    Jianqi An, Jinhua She, Huicong Chen, Min Wu

    Abstract

    Metallurgical industry is one of the most important industrial processes, which mainly consists of coking process, sintering process, ironmaking process, and casting and rolling process. All of the metallurgical processes are complex, multivariate and nonlinear systems with large time-delay. Some chemical or physical mechanisms are even not clear and uncertain. It is difficult to establish the models, design the controllers, devise the scheduling and optimization strategies, and make the operation decisions by the conventional mechanism-based methods. Nevertheless, these processes work continuously and repetitively, which produces large amounts of data, and consists of lots of knowledge and expert experiences. In the last decade, evolutionary computation and artificial intelligence (ECAI) began to be widely used in metallurgical industry and many good results were reported. This letter demonstrates how the development of ECAI impacts the metallurgical industry by analyzing some good applications of the ECAI in typical metallurgical processes and discusses the future development trends and challenges of the applications of the ECAI in metallurgical industries.

11. Evolutionary Computation and Artificial Intelligence for Business Transactions

    Apostolos Gotsias

    Abstract

    Business transactions are at the core of economic analysis as well as a prime area for research in Evolutionary Computation and for implementing scenarios in AI settings. The paper’s focus is on the issue of synchronizing activities and coordinating the agents of the firm in a supply-chain environment. After briefly discussing the general coordination model, the Hourglass model, we present a mathematical model for achieving coordination inside the firm and show how the agents’ activities are coordinated in a department, as well as across departments. The coordination model specifies the synchronization conditions by considering message travel times and product/support operational requirements. The conditions for achieving coordination and the relationships between operational and support departments are an original contribution in the economics of the firm. In the final part of the paper, we indicate how the coordination model’s results can be utilized in a dedicated AI environment for studying economic relations among firm/market participants.

12. Proposal on Mutual Cooperation Between Simulation Research and Field Research in Archaeology

    Fumihiro Sakahira

    Abstract

    In this paper, we propose a methodology to collaborate the research using agent simulation and the research using conventional method (field research) in archaeology. The main stream researches using Agent-Based Simulation (ABS) are unidirectional cooperation with the results of field research as input and the results of simulation research as output. In our proposed method, by presenting the hypothesis verification method from ABS result, the simulation result can become the input of field research. As an application example of proposed method, we discuss the problem of whether native Jomon people or Chinese-Korean immigrants played the major role of agricultural culture in Yayoi period by ABS.

13. Beyond Educational Policy Making

    With Agent-Based Simulation Atsushi Yoshikawa, Satoshi Takahashi

    Abstract

    In recent years, formulation of educational policy has come to be based on data. That data, however, can turn out to be difficult to access, or mixed with so much noise interfering with education policy formulation, that it cannot be used directly for policy making. To address this issue, an increasing number of attempts to contribute to policy formulation have been made using agent-based simulation (ABS). In the majority of research, ABS is used in the ex post facto analysis of why educational policy has not been effective. In this paper, case studies show that by incorporating ABS into the policy formulation process, the risk of failure can be reduced. By illustrating the relationships between model level, stage of educational policy formulation and the output scenarios of ABS, it is possible to determine which types of risks can be reduced. This paper presents ABS description levels, and discusses risks that both can and cannot be expressed using ABS. We show two ways to use ABS for educational policy making by identifying risks that can be reduced and risks that cannot be dealt with by ABS.

14. Backmatter