مدل سازی فرایندی پویا برای BPR : یک روش شبیه سازی کامپیوتری

Business Process Redesign (BPR) projects are considered to be high risk due to their high management complexity, enterprise-wide impact, and steep project cost. This paper describes a technique that should reduce that risk by developing a systematic process redesign method that we call Dynamic Process Modeling (DPM). It integrates customer-oriented process modeling with computerized visual process simulation to promote better understanding of the required process and determine its performance through simulation of the proposed redesign alternatives prior to selection and implementation. We compare DPM with four other implementation-level process modeling methods using eight criteria to demonstrate its effectiveness in a real-world hospital BPR case.

Business organizations in the 1990s are facing anever-increasing uncertainty and unprecedented volatilityof the external environment. Increased competitionhas led many organizations to the "fundamentalrethinking and radical redesign of their age-old business processes" called Business Process Redesign(BPR) [4], or Business Reengineering (BR) [ 10]. Suchefforts tend to be complex, have larger enterprise-wideimpact, and steeper project cost than the traditional ISdevelopment projects today. Despite isolated successstories at several firms [12], many organizations haveencountered serious problems during their BPR implementationswith widely mixed results [8, 9]. So BPRshould always be considered a high-risk project fromthe firm's perspective.In this paper, we try to show how to reduce the riskof BPR projects by two means: process modeling andcomputerized simulation. The first is a technique forunderstanding, representing, and, when necessary,redesigning thefundamental business processes. Lackof a disciplined method to model business processeshas been a problem in many BPR efforts [2]. An idealprocess modeling method for BPR would provide asimple but expressive modeling mechanism thatreflects the customer orientation and cross-functionalnature of BPR [3]. One such method, the Event-Process Chain (EPC) modeling method [14], wasextended here. By sacrificing the conceptual-levelclarity and readability of the EPC model, the ExtendedEPC (EEPC) model can represent the more detailed implementation-level process dynamics by supportingadditional modeling elements. Despite the massive resource investment and itsenterprise-wide impact, many firms have been quickto 'drastically redesign' their major business processeswithout a thorough analysis of the impact of thechange. All too often, instead of 'dramatic performanceimprovement' they end up with impractical andirrevocable process change that is hard to justify.Computerized process simulation may help to resolvesuch a problem. By simulating the expected performanceof the proposed alternatives, it will be possibleto choose the best [15, 20, 22].Finally, by integrating EEPC modeling and computerizedprocess simulation techniques with performance-based guidelines, an innovative BPRmethodology called Dynamic Process Modeling(DPM) is introduced and a real-world hospital caseused as an illustration.

In this research, we suggested a dynamic process modeling (DPM) method to support firms interested in redesigning their business processes with thehelp of a simulation technique. Such a modeling method enables us to model, not only the staticelements of the process, but also its dynamic aspects.The dynamic process model - EEPC - can help whilemodeling and redesigning the business processentirely. Animated simulation techniques based onthe model enabled us to perform evaluations. Andthe EEPC model, a customer-oriented cross-functionalprocess model, is easy to use in communicatingwith non-IS employees or managers, though the discrete-event simulation model is rather difficultfor the non-specialists to use.To demonstrate the practicality of the method,we applied it to a real hospital. As this reveals,the DPM method reduces the risk of BPR by providinga better understanding of the target businessprocess from the customer's viewpoint and showinghow performance is affected by the proposed alternatives.However, the hospital example we used oversimplifiesthe situation experienced in industry. First,although there are various satisfaction-related factorsin industry, we selected only three general factors(delay time, utilization, cycle time). In real BPRprojects, we need to consider each business-specificfactor. Second, the example did not deal with the cost issue. In a real project, we should compare eachalternative in terms of the trade-off between costand other tangible and intangible benefits. Third,process-related data are not easy to obtain. For example,the station's capacity is required for the simulation.But each station's capacity is frequently not fixedor is hard to calculate because a server is ofteninvolved in more than one process and it may needto be performed by more than one server. The best method is the use of existing corporate data. However,this is seldom possible.