THE MODELING AND SIMULATION OF A PICK AND PLACE COMPUTER-INTEGRATED MANUFACTURING (CIM) CELL

Citation
H. Bartlett et J. Harvey, THE MODELING AND SIMULATION OF A PICK AND PLACE COMPUTER-INTEGRATED MANUFACTURING (CIM) CELL, Computers in industry, 26(3), 1995, pp. 253-260
Citations number
11
Categorie Soggetti
Computer Application, Chemistry & Engineering","Computer Science Interdisciplinary Applications
Journal title
ISSN journal
01663615
Volume
26
Issue
3
Year of publication
1995
Pages
253 - 260
Database
ISI
SICI code
0166-3615(1995)26:3<253:TMASOA>2.0.ZU;2-I
Abstract
This paper presents the work carried out on the modelling and simulati on of a pick and place CIM cell. The CIM cell was designed to manufact ure printed circuit boards (PCB) and consisted of four numerical contr olled machines, two conveyor systems and a number of electronic device s all connected to an Ethernet local area network (LAN). With many com panies across Europe [1] and the USA [2] now implementing Ethernet int o their manufacturing environment, it is important that for time-criti cal CIM systems there is sufficient capacity within the Ethernet proto col so that there is no loss in production and that bottlenecks within the cell do not occur due to long message delays. This paper pays par ticular attention to the information processing function, the network communication protocol Ethernet. This communication protocol, although very popular, can give rise to long message delays which is not condu cive to time critical CIM systems. In order to determine the maximum m essage delay which hitherto has not been considered has led to the dev elopment of a complete model of the pick and place CIM cell in which b ath the operational and information processing functions were modelled and simulated using two independent simulation tools, SIMAN and L.NET . Results from the complete model have shown that providing the maximu m message delay, within the pick and place CIM cell, is less than the smallest machine operating time, there is no loss in production and bo ttlenecks do not occur. The paper concludes that the approach adopted in this paper could be used for any CIM system and that determining th e maximum message delay in this way gives a very important design para meter particularly for time critical CIM systems.