Improving Process Descriptions in Research by Model-Based Analysis

2021

Applied Sciences

Self Cite

Model-based Systems Engineering (MBSE) approaches are a step forward in the evolution of computer-aided engineering, and yet, they often incorporate deficiencies that may jeopardize their practical utility and usability, as well as the validity of the resulting models. We demonstrate how a domain-specific modeling approach can relieve some hurdles in adopting MBSE, and how it can be used in tandem with a general-purpose modeling approach to augment and introduce rigor to models. Specifically, we demonstrate the consequences of theoretical issues that were previously identified in Object Process Methodology and suggest an approach to solve them. We use a generalized case-study—derived from extensive process modeling in both academia and industry—to show that a domain-specific model can significantly relax the user’s modeling effort. This demonstration is based on two quantitative metrics: the number of representational elements and available modeling tactics. We discuss the contribution of our approach to model quality, particularly with respect to its rigor and communicability.

Section: Modeling a Process With Dsm (Prove)mentioning

confidence: 99%

mentioning

confidence: 99%

Using Domain-Specific Models to Facilitate Model-Based Systems-Engineering: Development Process Design Modeling with OPM and PROVE

Shaked

2021

Applied Sciences

Self Cite

“…Davis and Sitaram [51] and Spillner [52] all noted similar characteristics. We have recorded some related observations from real projects in previous publications [53][54][55].…”

Section: Ontology Of Development Process Designmentioning

confidence: 99%

Requirements for Model-Based Development Process Design and Compliance of Standardized Models

Shaked

2021

Systems

Self Cite

The planning of system development efforts is crucial to the successful realization of projects. However, development planning typically lacks systematic, engineering discipline, and consequently risks project and business success. Model-based process design is a potential information systems approach to addressing the increasing complexity of such planning. We characterize the ontology of development process design, based on real-life observations and scientific publications. We then synthesize the required ontology with the desirable characteristics of models, and derive key requirements for model-based development process design. Next, these requirements are used to evaluate the adequacy of three prominent, standardized model-based process design approaches—BPMN, OPM and SPEM. The findings reveal that the surveyed models are a partial fit, and do not promote sound process design. Finally, by generalizing the categorical evaluation results, possible root causes for the identified inadequacies are proposed. A new model design, which should rely on the formulated requirements set, is called for, in pursuit of a wider adoption of model-based design paradigms and better information systems realization to support the development of complex systems.

“…This proposed abstract language allows for describing diverse models to understand the structure and behavior in a design context or a study. The benefit of models or theories (which are also models) for analyzing and comparing design methods has a lineage from before (e.g., Kroll et al, 2014;Reich et al, 2012;Shai et al, 2013;Sitton & Reich, 2019;Shaked & Reich, 2020). Nevertheless, in these studies, models were used to analyse one or two studies separately, to compare their genericity and inter-relationships, with formal models.…”

Section: Introductionmentioning

confidence: 99%

Mapping and Enhancing Design Studies With Psi Meta-Theoretic Design Framework

Subrahmanian

2021

Proc. Des. Soc.

Self Cite

Design research as a field has been studied from diverse perspectives starting from product inception to their disposal. The product of these studies includes knowledge, tools, methods, processes, frameworks, approaches, and theories. The contexts of these studies are innumerable. The unit of these studies varies from individuals to organizations, using a variety of theoretical tools and methods that have fragmented the field, making it difficult to understand the map of this corpus of knowledge across this diversity.In this paper, we propose a model-based approach that on the one hand, does not delve into the details of the design object itself, but on the other hand, unifies the description of design problem at another abstraction level. The use of this abstract framework allows for describing and comparing underlying models of published design studies using the same language to place them in the right context in which design takes place and to enable to inter-relate them, to understand the wholes and the parts of design studies.Patterns of successful studies could be generated and used by researchers to improve the design of new studies, understand the outcome of existing studies, and plan follow-up studies.