Adaptive Search and the Management of Logistic Systems : Base Models for Learning Agents

The management problems addressed in Adaptive Search and the Management of Logistic Systems stem from the recent changes in the logistics sector. Global competition and increasing consumer expectations are forcing industrial enterprises to reorganize their business processes to support better and more cost-efficient services. Quick and precise customer delivery processes have been recognized as the key to significant savings and growth for a company. In seeking solutions to the related problems of planning and control in today's logistic activities, the book is divided into two parts. Part I lays down the fundamentals of evolutionary adaptive systems before an architectural model for adaptive agents is developed. Part II is devoted to applications of these adaptive agents. By focusing on elementary yet complex problems, it is hoped to glean sufficient details of the real world. In its examination of the logistics network, the book employs a simple yet powerful framework of evolutionary search, allowing efficient support for reactive problem solving by the actors in the logistics network. Following modern terminology, we refer to evolutionary algorithms as learning or adaptive agents. Different from other software artifacts, which are also designated as agents in the current literature, these adaptive agents do not communicate directly by exchanging messages. Instead, while exploring the space of logistic operation, they communicate by mutually changing details of the common environment they perceive