This paper addresses the design of an integrated tool management system for¯exible machining facilities (FMFs). Modules with functions ranging from issuing tools according to a tooling strategy to diagnosing system operation have been developed and integrated around a centralized manufacturing database to guarantee streamlined manufacture. A review of the state-ofthe-art tools to model an automated tool management system (TMS) within¯exible machining facilities is presented ® rst, in order to establish the merits of diVerent design approaches. The approaches used in modelling, the structures and the functions of the modules are presented. Finally, a walkthrough example utilizing the design facility presented is given with an illustration of typical results. NOTATIONc transporter journey index from 1 to C journeys DCA dynamic cluster analysis EDD earliest due date i tool index from 1 to T tool types j job index from 1 to J jobs KBOA knowledge-based output analysis LPT longest processing time L i cutting time for tool i (min) m machine index from 1 to M machines MT throughput time MU m machine utilization P index set of pallets, pˆ1; . . . ; P o operation index from 1 to O RBTMSS rule-based tool management strategy selection S set-up time SIM 1;2 similarity between two cluster sets SPT shortest processing time ST i spent tools TI m minimum tool inventory TI s predicted tool inventory TMS tool management system TR tool requirements for a batch TR m minimum tool requirements TRP tool requirement planning TT tool traYc (tool movement) TTR total tool requirements TU transportation utilization T m available time for machine m, m 2 M X ojpmˆ1 if operation o of job j of pallet p is assigned machine m 0 otherwise Z ojpˆ1 if operation o of job j of pallet p is assigned 0 otherwise 1 INTRODUCTION

EXECUTIVESUMMARYThe management of tooling systems in cell-based flexible manufacturing systems is described. The use of cluster analysis in the management of cutting tools is presented. The role and functioning of a cluster based tool management strategy is described with respect to the management of the cutting tools, scheduling of the work flow and the number of captive tools. The research findings are described and supported by an example. A Rank Order Clustering (ROC) algorithm was used, selected for its simplicity and rapid implementation, to cluster tools. The ROC algorithm is applied to cluster tools. These clusters are scheduled on work stations based on part (the primary priority) and cluster (secondary) priority management rules. This provides the dynamics of managing the change of tool clusters as well as providing for the case where resident tool clusters are desired. A spreadsheet-based computer assisted tool clustering model, intended for industrial usage, is described. The clustering model reduces the requirement for robust advance schedule generation and encourages the adoption of smaller tool magazine capacities. The cluster based model advances beyond just tool kitting to provide rapid on-line generation of tooling configurations where part mix and production ratios may be varied almost instantaneously on a spreadsheet and where rigidity in scheduling may be relaxed by allowing a tool cluster to attract any part in the cluster set. The paper concludes with a brief description of on-going research effort in applying cluster analysis to cell tool management.