Thresholds for error probability measures of business process models

Abstract

The quality of conceptual business process models is highly relevant for the design of corresponding information systems. In particular, a precise measurement of model characteristics can be beneficial from a business perspective, helping to save costs thanks to early error detection. This is just as true from a software engineering point of view. In this latter case, models facilitate stakeholder communication and software system design. Research has investigated several proposals as regards measures for business process models, from a rather correlational perspective. This is helpful for understanding, for example size and complexity as general driving forces of error probability. Yet, design decisions usually have to build on thresholds, which can reliably indicate that a certain counter-action has to be taken. This cannot be achieved only by providing measures; it requires a systematic identification of effective and meaningful thresholds. In this paper, we derive thresholds for a set of structural measures for predicting errors in conceptual process models. To this end, we use a collection of 2,000 business process models from practice as a means of determining thresholds, applying an adaptation of the ROC curves method. Furthermore, an extensive validation of the derived thresholds was conducted by using 429 EPC models from an Australian financial institution. Finally, significant thresholds were adapted to refine existing modeling guidelines in a quantitative way.

Empirical investigation of the usefulness of Gateway constructs in process models

“…First, we selected appropriate complexity metrics based on the set of metrics defined by Mendling (2008). The most important classes of metrics apply to the size of the models (most notably, arcs and nodes), and the connections within the models (most notably average connector degree) (Mendling et al, 2012a). Next, we defined three levels of complexity, viz., low, average and high.…”

A systematic literature review of studies on business process modeling quality

“…understandability) of the process models. For example, several authors agreed that an increase in size of a model appears to have a negative impact on its pragmatic quality [6,[39][40][41][42][43]. Some considerations are available on when a process model would have to be split up into subprocesses to decrease its size.…”

“…It has been recommended based on empirical findings that process models with more than 50 elements should be decomposed [6]. Another study proposes to decompose the model once it has more than 31 elements [43] based on a threshold definition. Depending on the process modeling language the amount of activities can vary for the same amount of elements [44].…”

Managing Structural and Textual Quality of Business Process Models

“…The work reported in [42] uses logistic regression and error probability as a dependent variable. Logistic regression is a statistical model for estimating the probability of binary choices (error or no error in this case) [43].…”

“…The best threshold can then be found based on sensitivity and specificity values with: sensitivity = true positives(TP) rate = T P P , specif icity = 1 − false positives(FP) rate = 1−F P P . Using this approach, several guidelines of the 7PMG could be refined in [42]. Table 1 provides an overview of the results showing, among others, that process models with more than 30 nodes should be decomposed.…”