Business process re-engineering (BPR) for new business processes or change of existing processes is not an ad-hoc activity. It requires a good project management approach to coordinate and decide on a range of factors in a structured manner [1]. However not all BPR projects are successful. Whittaker [2] conducted a large survey of companies and concluded that 45% of improvement projects did not deliver the expected benefits.

A successful BPR project depends on the capability of the company implementing it. A proper analysis of known success and failure capability factors can reduce the risk of failure and increase the likelihood of success. The structured process improvement framework proposed by Rohleder et al. [3] had the potential to simplify the change implementation process and avoid failures but their focus was on innovation which they acknowledged to have a high failure rate.

This paper outlines a risk assessment method to assist organisations re-engineering their business processes in order to achieve the expected performance gains. The innovation of this method is the ability to estimate the probability of meeting a desired target performance level at various points in time. This knowledge is most critical to decision makers to a) compare different BPR projects and execute the project with the highest probability of success over time, and b) form the basis for adjusting the scope and performance targets of BPR projects to achieve a realistic chance of success.

Performance measures are characterized through metrics. Consolidated sets of performance indicators are used by company management to ensure attainment of their objectives and goals linked to strategy and value creation for the company. Prandini et al. [4] specified the complexity metrics that were relevant in the operations of future air traffic management system. Marin-Garcia and Poveda [5] reviewed the success of a medium sized company implementing change suggestions from the employees. The subjective approach evaluating employee suggestions had a competition effect among multi-disciplinary teams in the company. Bourne [6] described three phases of performance measurement system implementation and pointed out that senior management commitment was crucial to the success of the project. Bourne et al. [7] investigated the performance measurement system design of ten companies. The study revealed two main drivers for success which were closely related to the perceived benefits of top management.

The problem of setting performance measures has been studied extensively in the manufacturing sector. Ahmed et al. [8] observed that additional tangible and intangible measures were required to generate sufficient information representing different aspects of performance evaluation. Lillis [9] studied the effectiveness of profit centre strategy where performance dimensions relating to manufacturing efficiency and customer responsiveness was found problematic. Gomes et al. [10] determined predictive value as a primary factor in the choice and level of usage of performance measures by managers with ease of data acquisition being subordinate in importance.

In a comparison of eight performance measurement systems (PMS), Garengo et al. [11] found a trend over time that small and medium enterprises changed from focus on strategy alignment alone to a focus on stakeholder satisfaction together with strategy alignment. Performance indicators used to measure process improvement should be linked to the performance measures otherwise the importance and relevance of the measured outcome of the project could be questioned at the outset [12]. There is a lack of research to link performance metrics to capability level of the organization working on BPR projects.

A thorough risk assessment provides benefit to decision makers coordinating changes in BPR to minimise potential failure and reduce effect of undesirable events. In Figure 1, the expected performance gain P_exp is aimed to be achieved at time t_j from the start point t₀ and performance P₀. Without prior knowledge of the risk level, the decision maker anticipates that curve 1 applies; that is a gain from P₀ to P_exp is obtained and maintained in the longer-term. However, there may be circumstances pushing the project to scenarios 2 to 6, each having

different probabilities. It is clear the probability of achieving the initially expected gain P_exp is somewhat over optimistic with a probability of success less than five percent. Instead, if the expected performance is adjusted to P_exp1, the chance of achieving the expected performance is close to half.

Lowering expectation levels is a decision that can be made conservatively and provides the short-term benefit of improving the momentum of improvement. Alternatively on being made aware the probability of success is lower than expected a decision can be made to expand existing capability to cover the gap.

Performance gains are anticipated from a BPR project. These performance gains are measured by a range of individual performance indicators P = { P ₁, P ₂, …, P _m} where the P_i (i = 1, …, m). For the convenience of decision making, the current performance indicators P_cur is a management defined function. In practice, a linear weighted sum is used as shown in Equation (2). Effort required to achieve higher levels of performance often involves greater complexity and cost. A performance effectiveness function k_P can be defined as a measure of the perceived effort required to overcome difficulty and complexity to achieve a particular level of performance from the current level P₀. Each of these performance outcomes are measured from a performance scale ranging from success (highest score) to failure (lowest score).

The performance of a BPR project can vary between two extremes:

§ A theoretical performance level P_th that can be attained if all capability elements are applied in a perfect manner.

§ The capability elements are not effective and no improvement of productivity has been recorded. This performance level is represented as P₀. If the project is not implemented, P_cur = P₀.

However, the problem of finding the intermediate relationships between k_P and P is mixed with human judgement. One method is to solicit the opinion of the stakeholders involved in the operations improvement project. The Analytic Hierarchy Process (AHP) is applied to consolidate the vector of priority (VP) over the range of P₀ to P_th. To assist in making consistent decisions, Equations (3), (4) and (5) are applied in the judgment matrix. For any j = 1 to th (i.e. th = 5 in this illustration and * is the wild card for the row or column elements),

A pairwise comparison judgement process on θ_kj with a scale of 1 to 9 generates intermediate values of VP_j. The performance effectiveness factor k_P is obtained over the range 0 to 1 by Equation (6).

Intuitively it is reasonable to expect that having a higher level of capability, the organisation can achieve greater levels of BPR performance gain closer to the theoretical maximum. These curves are obtained by estimating the difficulty levels k_P in past improvement projects against the increase in performance P on a particular S_C. The derivation of this function is based on published data [14] [15].

The goal of this research is to predict future performance gain of an operations improvement project based on an estimate of capability for that project. Assuming an estimated S_C of 70 for the project (Figure 3), and by combining the k_P/S_C and k_P/P curves a prediction is made. From Figure 3, with 90% confidence, the performance level at a future time will be in the range [2.5, 3.7], with an average performance level of 3.2. If the target performance is set at 3.7, it is then up to the decision maker to decide whether a risk of 10% failure is acceptable for the PBR project.

A risk assessment model has been proposed to assess the likelihood of meeting performance gain targets from BPR projects undertaken by companies. The method indicates to companies when and where their capabilities need to be adjusted to maximize their possible outcomes. This alleviates the problem that organizations have when they embark on BPR initiatives with an organizational setup that limits their chance of success.

An advantage for companies using the S_C model is their improvement ambition level can be set more realistically based on knowledge of their estimated capability. The adjusted performance targets are better matched to organizational capability which in turn means a higher hit-rate of successful BPR projects. More success tends to provide positive feedback setting the seeds for sustained continuous improvement. Setting the bar too high too early can demotivate a BPR programme and kill it before it even gets going.

Boyd A. Nicholds,John P. T. Mo, (2015) Risk Assessment of Business Process Re-Engineering Projects. Open Journal of Social Sciences,03,30-34. doi: 10.4236/jss.2015.33007