Highly Efficient AGV Transportation System Management Using Agent Cooperation and Container Storage Planning Satoshi Hoshino and Jun Ota Dept. of Precision Engineering, School of Engineering The University of Tokyo Bunkyo-ku, Tokyo 113-8656, Japan {hosino, ota}@prince.pe.u-tokyo.ac.jp Akiko Shinozaki and Hideki Hashimoto Mitsubishi Heavy Industries, LTD. Sagamihara-shi, Kanagawa 229-1193, Japan Abstract— The development of a highly efficient management methodology for an Automated Container Terminal (ACT) poses a problem for port authorities. The focus here is on a trans- portation system for an Automated Guided Vehicle (AGV) for an ACT. In this paper, we design the detailed management models, i.e., agent cooperation and container storage planning for the transportation system. Then, we optimally design systems that are constructed with the use of the designed management models. Comparisons of the systems are made to evaluate cost effectiveness based on the total construction cost and validity of the management models. Finally, a proposal is made for the most efficient management system. Index Terms— AGV transportation system, system manage- ment, agent cooperation, container storage planning. I. I NTRODUCTION The increasing number of container shipments has caused higher demands on seaport container terminals, container logistics, and management, as well as on technical equipment. In this regard, concern is increasing regarding Automated Container Terminals (ACTs) in harbors [1] [2]. In this study, we construct an automated transportation system with the use of an Automated Guided Vehicle (AGV) as a transportation agent. For the construction of the AGV transportation system, there are three problems that need to be overcome. They are presented in the following. (I) Optimal design of the transportation system (II) Comparison, evaluation, and analysis of the trans- portation systems (III) Highly efficient management methodology for the transportation system Problem (I) represents a design methodology that is related to the challenge: deriving the combinatorial optimal design parameters, i.e., the minimum number of agents that satisfy a demand. For this problem, we have proposed a hybrid design methodology with the use of queuing network theory and a simulation model [3]. On the other hand, we need to carefully identify the most efficient system when several systems, such as vertical and horizontal transportation systems, are being considered (problem II). For this problem, some studies have compared, evaluated, and analyzed the cost effectiveness based on the transportation time and total construction cost [4] [5] [6]. From the results of conventional studies, it is clear that the horizontal AGV transportation system is more cost-effective than the well-known vertical AGV transportation system under a cost model [5]. Therefore, in this paper, we deal Quay area Container ship QCC AGV (Loaded) AGV (Empty) Transportation area Transportation area Container storage area RTGC 1st location 2nd location 3rd location nth location 350 [m] 110 [m] 320 [m] Work path 32 [m] Stopping AGV Fig. 1. Horizontal AGV transportation system (top view) with the specific horizontal AGV transportation system, as shown in Fig.1. However, since a Rubber-Tired Gantry Crane (RTGC) that operates in the container storage area is also an important and flexible agent in addition to the AGV in the horizontal transportation system, a methodology for the management of such a complex system is required. For this paper, we construct an AGV transportation system while considering this problem (III). In this regard, since there are various types of agents, i.e., heterogeneous multi- agent operating in a transportation system, we first design the detailed system management models that would tackle this problem (III). For the modeling, we need to take into consideration: when, where to, and how the containers should be transported and stored by the agents. Therefore, two challenging points are important for a highly efficient system management: agent cooperation and container storage plan- ning. For the modeled system, we derive the combinatorial optimal design parameters using the proposed design method- ology [3]. After that, we compare the cost effectiveness of the designed systems based on the total construction costs, and we evaluate the validity of the management models. Finally, we overcome the problem (III) and are now proposing the most highly efficient management model.