توسعه‌ی یک راهبرد تصمیم‌گیری برای بهبود کارایی BPR

To support the efficient appraisal of and selection from a list of generic business process improvement principles, this paper proposes a strategy for the implementation of business process redesign (BPR). Its backbone is formed by the analytic hierarchy process (AHP) multicriteria method and our earlier research into the popularity and impact of a set of redesign “best practices”. Using AHP, we derive a classification of most suitable directions for a particular process to be redesigned. Criteria such as the popularity, the impact, the goals and the risks of BPR implementation are taken into account. A case study is included to demonstrate the method’s feasibility and effectiveness.

One of the highlights in a large survey among senior business managers is that business process redesign (BPR) is almost as popular again as it was in the beginning of the 1990s (Rigby & Bilodeau, 2005). Despite the continued interest in this approach to rethink existing process structures considering the opportunities that IT provides, few analytical tools exist to support the actual redesign of a business process (Nissen, 1988). The aim of the work as presented in this paper is to develop a tool that supports the decision-making process practitioners apply to come up with a new, improved plan for a business process. This aim links up with a more general observation that BPR often does not lead to the desired results, because it is a time-consuming and costly affair with unpredictable results. It has been argued that there is a clear need to improve the redesign process itself (Hofacker and Vetschera, 2001, Nissen, 1988 and Reijers, 2003). The goal of the decision-making tool that is described in this paper is to: (i) increase the efficiency of the redesign process itself, and (ii) to lead to a more systematic evaluation of improvement opportunities. In earlier work (Limam Mansar and Reijers, 2005, Limam Mansar and Reijers, 2007 and Reijers and Limam Mansar, 2005), we published on our efforts to attain the second goal through the identification, validation, and practical use of a set of so-called “best practices”. In this context, a best practice is a general heuristic derived from earlier successful encounters to improve process performance, which may need skilful adaptation to be applied in a concrete setting. For example, instead of using a paper file which favors processing in a sequential way (i.e. the physical document is passed from one executor to the other), the use of an electronic file may be considered to speed up the work, as people can work concurrently on their own electronic copies. The proposed set of best practices may be used to structure the redesign sessions with business professionals, as we did, for example, for the redesign of an intake procedure in a mental health-care setting (Jansen-Vullers & Reijers, 2006). Each best practice was considered by all participants on its applicability and subsequently subjected to a more thorough performance evaluation by simulating the process models. But even though this structured approach improves upon the often intuitive way that BPR is carried out, it remains problematic in the sense that such an approach requires considerable time and efforts from all participants to carry out the project. The described tool in this paper still fosters the systematic breadth of considering a set of redesign best practices, but it also addresses the efficiency of the BPR process by efficiently classifying a set of most appropriate best practices for a specific case. Such a result may serve as a “kick-start” for the redesign team involved, speeding up the redesign process. There have been other contributions in this field where mainly artificial intelligence algorithms have been used. Case-based reasoning and inference rules are examples of such approaches (see e.g. Min, Kim, Kim, Min, & Ku, 1996). However, the majority of these contributions require the gathering of a large set of successful cases or address only specific processes for a given industrial or service sector. An exception is the work of Nissen (Nissen, 1988) that aims to detect weaknesses in a given process design by using various metrics and dedicated algorithms. Although the aim of this work is comparable to ours, the approach is completely different, as will be discussed in our related work section (see Section 7). In the remainder of this paper, Section 2 will give the necessary background for this paper in the form of an overview of our earlier work. Section 3 gives a high-level description and contribution of our tool and specifically how it may help to improve upon common design practice. Section 4 deals with introducing the different aspects or criteria that should be taken into account when deciding which best practice should be implemented in a concrete situation. Section 5 introduces AHP as the multicriteria decision-making method chosen for this study and builds up the strategy for the implementation of BPR using AHP. Section 6 applies our findings to the case study of a Dutch municipality. Section 7 is a review of related work. Finally, Section 8 provides our conclusions and future work.

This paper presented a decision-making method based on AHP to support practitioners in the field of BPR to choose appropriate best practices to enhance processes. It is important to emphasize that this work supports business process redesign where process improvement is the goal as opposed to a clean slate approach to redesign. In that context, the main contribution of this paper is twofold. Firstly, it presents a new approach to business process redesign (refer to Section 3 and Fig. 1). This approach aims at shortening the time practitioners will spend discussing the usefulness of best practices and at providing them with a much clearer appreciation of best practices importance and impact. Secondly, it synthesizes all important criteria for selecting best practices, introducing the goal and risks of the examined projects. This forces the redesign team to reflect on the conditions in which the redesign is conducted. Secondly, the presented parameterization and application of the AHP algorithm to select best practices for a specific redesign project is a further step towards the development of a decision-making strategy for BPR. Based on earlier work of ourselves and others, we came up with substantiated comparisons and weights of the tables that are used in the appraisal of AHP. Although some values were derived in previous papers (Limam Mansar and Reijers, 2005, Limam Mansar and Reijers, 2007, Mansar et al., 2006 and Reijers and Limam Mansar, 2005), they were not presented or organized in such a way that it would clearly indicate how they can be used by practitioners. Despite the well-known disadvantage of AHP in terms of effort required to rate criteria and options, the rating is really restricted to defining the relative weights for the indicators of the goal and the risk criteria only (equivalent to filling the values of Fig. 6 and Fig. 7). Also, the use of AHP in the presented method should be seen in the light of a more rational approach towards BPR, attaining the advantages of a more systematic search for redesign opportunities but more efficiently than by using simulation. Our method has been implemented in Java 2 (JBuilder), compatible with the JDK standard 4.0 and higher. For now, data in the form of different tables is entered in text files and used by the algorithm. The results are then generated and displayed visually on a computer screen. The algorithm takes virtually no time to make the computations for the size of data that we considered here. Clearly, our implementation is still a prototype and needs enhancement of its user interface to be fully used in a business environment. Currently, contacts with a Dutch investment bank are exploited to enhance the tool with respect to this issue. Using a case study in this paper, we showed the feasibility and potential of the approach. We compared and discussed its outcome to the real application, where simulation was used. The presented method allows the user to identify potentially interesting best practices. Although AHP’s output is a ranking of best practices, we do not think the exact ranking is important. What should be considered is merely a preferred set of best practices as opposed to all best practices or the best practice. Indeed, in none of the redesign initiatives the authors have been involved in merely one best practice was singled out. Discussions always resulted in indicating a preferred set that later was tested or applied. With the presented method, the best practices’ relevance is addressed through a “high-level” assessment. It considers factors such as the redesign project’s risks and goals and the characteristics of the best practices, for example their impact on the processes’ performance indicators. In this respect, it is different from the work of Nissen (Nissen, 1988) that takes a “low-level” approach. This work considers the structure of a given process in much detail (e.g. length of the paths, number of steps, etc.) to identify improvement opportunities. It would be interesting to see whether these approaches could actually be used together or even merged and by doing so, take another step in moving redesign from art to science. Future research should also focus on further validating the values assigned informally in this paper, namely the relative importance of the criteria and of the impact indicators as well as the relationship between the risk’s and goal’s indicators values and the best practices