Dynamic information support system for the process of prefabricated member installation using RFID technology and multi-agent system

Nowadays, high expectation of efficiency and effectiveness encourages the prefabrication usage in housing construction industry. However, the expected benefits cannot always be obtained. One of the reasons is the inadequate skill and experience of personnel for normal situation and changed situation. The skills and experiences are required for some tasks in the normal situation, although personnel can follow the static information which is created based on a planned scenario. In the changed situation, the static information may become inappropriate. The appropriate information must be updated and provided to personnel using the current information and conditions. Otherwise, the personnel cannot perceive the actual situation and select the countermeasures properly. Consequently, undesired productivity and mistakes tend to occur. Thus, this research aims to develop a support system for improving productivity and for reducing the number of mistakes in the process of prefabricated member installation by identifying available resources, collecting actual conditions, managing knowledge, generating alternative methods, and providing appropriate information. The system is developed by the integration of Radio Frequency Identification (RFID) technology, a multi-agent system, and a database using Microsoft Visual Studio, Java Agent Development Framework (JADE), and Microsoft SQL Server. The RFID technology is employed for identifying resources. The multi-agent system consists of the following seven agents: (1) information providing agent, (2) general information agent, (3) specific information agent, (4) actual condition recording agent, (5) alternative generating agent, (6) quality assurance and control agent, and (7) evaluation agent. Additional five user interface agents are included in a user interface container. The database stores data such as inspection results, and other related information and knowledge. To verify the proposed methodology, a prototype is being implemented in a personal digital assistant (PDA) with a RFID reader and a laptop PC. The system will be tested in a real prefabricated housing construction site.