No abstract

Business process models are widely used for purposes such as information systems analysis, improving operational efficiency, modeling supply-chains, and business process reengineering. A critical aspect of process representation involves a choice among alternative or parallel routes. Such choices are usually represented in process models by routing structures that appear as "split" and "merge" nodes. However, evidence indicates that modelers face difficulties representing routing options correctly. Clearly, errors in representing routing options might negatively affect the effective use of business process models. We suggest that this difficulty can be mitigated by providing process modelers with a catalog of routing possibilities described in terms that are meaningful to analysts. Based on theoretical considerations, we develop such a catalog and demonstrate that its entries have business meaning and that it is complete with respect to a defined scope of process behaviors that do not depend on resources or on software features. The catalog includes some routing cases not previously recognized. We tested experimentally the catalog in helping subjects understand process behavior. The findings demonstrate that the catalog helps modelers understand and conceptualize process behavior and that the likely reasons are its completeness and the practical terms used to describe its entries. . 1 An exception is the domain-specific language Picture (Becker et al., 2010). Picture uses building blocks where decisions are embedded in activities (without splits). This is aimed at avoiding multiple conflicting representations of a situation, and to facilitate automated model analysis.process behavior. We then discuss the significance of the results, summarize the work, and suggest further research. Cognitive Aspects of Process ModelingEmpirical observations (Pinggera et al., 2012) have indicated that process modeling involves three phases. The first is comprehension, in which the modeler develops an understanding of the represented domain. The second, modeling, occurs when this understanding is transformed into modeling constructs. Thirdly, reconciliation occurs when model elements are reconciled, moved, and renamed, to improve appearance and clarity. These three phases are repeated, each iteration relating to a chunk of the model. Iterative chunking has been addressed by cognitive problem-solving theories (Newell and Simon, 1972), which suggest that tasks are typically addressed in smaller parts (chunks) due to limitations in working memory (Miller, 1956).We consider the construction of a process model to represent a given domain behavior as a problem-solving task, and the model as the solution. We focus on the comprehension phase, when the modeler develops a domain understanding. The domain understanding will be mapped into constructs of a modeling language.According to Newell and Simon (1972), the problem-solver formulates a mental model of the problem, and uses it to reason about the solution and to apply solution procedures....

No abstract

A common way to represent organizational domains is the use of business process models. A Workflow-net (WF-net) is an application of Petri Nets (with additional rules) that model business process behavior. However, the use of WF-nets to model business processes has some shortcomings. In particular, no rules exist beyond the general constraints of WF-nets to guide the mapping of an actual process into a net. Syntactically correct WF-nets may provide meaningful models of how organizations conduct their business processes. Moreover, the processes represented by these nets may not be feasible to execute or reach their business goals when executed. In this paper, the authors propose a set of rules for mapping the domain in which a process operates into a WF-net, which they derived by attaching ontological semantics to WF-nets. The rules guide the construction of WF-nets, which are meaningful in that their nodes and transitions are directly related to the modeled (business) domains. Furthermore, the proposed semantics imposes on the process models constraints that guide the development of valid process models, namely, models that assure that the process can accomplish its goal when executed.

Syntactically correct process models are not necessarily meaningful or represent processes that are feasible to execute. Specifically, when executed, the modeled processes might not be guaranteed to reach their goals. We propose that assigning ontological semantics to process modeling constructs can result in more meaningful models. Furthermore, the ontological semantics can impose constraints on the allowed process models which in turn can provide rules for developing process models. In particular, such models can be designed to be valid in the senses that the process can accomplish its goal when executed. We demonstrate this approach for Petri Net based process models.