Abstract – A Reconfigurable Manufacturing System (RMS) is having potential for quick adjustment in its production capacity and functionality. These RMSs are considered to be one of the technologies enabling shorter lead times, more product variants, fluctuating volumes of products, shorter product life cycles and lower cost. This paper presents the performance modeling of a hybrid system comprising of dedicated and reconfigurable machine tools simultaneously. The modeling focuses on handling of exceptional situations. One such handling situation arises when there is a breakdown of any resource like machine within the system. The dedicated machines are subjected to random failures and once there is the failure, the jobs are transferred to the reconfigurable machines after suitable reconfiguration to carry out the sequential operation required on the job. Two different production scenarios, first, operation of the system over some predetermined period of time and, secondly, operation of the system for a fixed quantity of products were modeled using ARENA ® discrete event simulation software. Three different dispatching strategies for reactive scheduling of parts based on total average waiting time in queue, total downtime of the failed machine and capacity of the buffer in between the machines were analyzed. The performance of the system was analyzed in terms of productivity, congestion and make-span time. Finally, the results were analyzed and were discussed in the light of previous researches carried out on the topic. Keywords - Reconfigurable Manufacturing System, scheduling, Performance evaluation I. INTRODUCTION An RMS offers ability to reconfigure hardware and control resources in order to quickly adjust production capacity and functionality in response to changes in the market conditions. These RMSs are considered to be one of the technologies enabling shorter lead times, more product variants, fluctuating volumes of products, shorter product life cycles and lower cost. All these product characteristics are responsible for a manufacturer to design its manufacturing system which can operate satisfactorily not only under stable and dynamic market scenario but also under static and dynamic conditions within the manufacturing plant as well. Some of the dynamic scenarios which may arise in a manufacturing system include worker absenteeism, machine failure, raw material stock out situation, previous backlogs, sudden change in the demand pattern and many more. An RMS is well equipped to meet these challenges. An RMS is basically an arrangement of CNC machines, dedicated machines, and reconfigurable machines tools (RMTs) to carry out various operations required on a variety of products. These RMTs are modular machines with flexible structure that allows changes of its resources like adding or removing axes of motions. The flexibility of RMS can be at system level in which the emphasis is to design the system configuration which can optimize reliability, quality and cost parameters [1]. Reconfigurability at machine level takes into account the design and analysis of RMTs along with their controllers and the ramp-up issues to address the problem of reducing the set up time of the system and to improve the make span time of the products while switching over from one production run to another [1]. The term “exception handling” can be defined as any situation under the dynamic operation state of the system in which the raw material, semi finished products or the finished products are to be transferred to other machines or resources which was not planned previously at the process planning stage. One of the most common causes of such handling is the unexpected breakdown of any resource like machine. Other exceptions on the shop floor may include changes in job priorities, dynamic introduction of new jobs, order cancellations, increase in job arrival rate, changes in product mix and rework due to quality inspection. These exceptions are generally unpredictable in nature. However, sometimes planned events like scheduled maintenance may also be treated as exceptions as they interfere with the production process. At times, the problem of exception handling becomes all the more important as it may cease the complete production process. Thus, there is a need to adopt some strategy to deal with the problem of exception handling. The basic objectives of exception handling policies may include (1) reduction of work-in-process inventory (2) balanced utilization of resources (3) reduction in average waiting times of different products in different queues and buffers (4) reduction in the make span time of different products (5) maintaining the required throughput, and (6) no significant increase in the cost of the products. The three main scheduling approaches for a dynamic and uncertain production system which are found in literature are: predictable scheduling, pure reactive scheduling and predictive-reactive scheduling [2]. Predictable scheduling focuses on the documentation and implementation of such strategies which consider some of the known events before, which may occur within the system while designing and operating such Performance Modeling of Reconfigurable Manufacturing System for Different Dispatching Strategies under Exception F. Hasan, P.K. Jain, D. Kumar Department of Mechanical & Industrial Engineering Indian Institute of Technology Roorkee Roorkee-247667, Uttrakhand, (Email: faisalhasan123@rediffmail.com) 978-1-4673-2945-3/12/$31.00 ©2012 IEEE 593