GoBIS: An integrated framework to analyse the goal and business process perspectives in information systems
Introduction
Organisations are aware of the importance of evolving to keep pace with changes in the market, technology, environment, law, etc. [1]. As a result, continuous improvement and reengineering have become common practices in information system (IS) engineering. Understanding organisations and their needs for change often requires several interrelated perspectives [2], [3]. The IS engineering community has contributed a number of modelling languages that are typically oriented towards a specific perspective, requiring approaches to their integration [4].
In this paper, we focus on extending a business process perspective with intentional aspects of organisations. Business process modelling languages provide primitives to specify work practice (i.e., activities, temporal constraints, and resources). Despite the fact that processes are widely accepted as a means to achieve organisational goals [5], process models give little attention to the strategic dimension [6]. The analysis, prioritisation, and selection of organisational strategies are the scope of intentional modelling languages, which focus on the business roles, their goals, and their relationships.
Business processes and goals are intrinsically interdependent [7] and several works provide detailed arguments in favour of combining both perspectives: (i) an integrated approach allows understanding of the motivation for processes [6]; (ii) in a non-integrated approach, goals may be used to guide process design [8]; (iii) traceability is enhanced, which is necessary for enterprise management [9] and facilitates the sustainability of organisations [10]; (iv) integration also helps in identifying cross-functional interdependencies during business change management by supporting the identification of the goals that motivate change and the analysis of their impact on processes [8], [10], [11].
The GoBIS framework (Goal and Business Process Perspectives for Information System Analysis) pursues this aim by integrating a goal-oriented and a business process-oriented modelling language. There are several criteria that one would expect from modelling language integration. For the framework definition, we highlight the following: (i) the languages to combine need to be formally described; (ii) the integration itself should be well founded in theory; (iii) it should clarify the scenarios where the integrated approach can be applied and provide some scenario-dependent guidelines; (iv) it should provide tool support; (v) empirical studies should demonstrate some benefits to potential adopters. These criteria guided our research. A comparative review (see Section 2) reveals that proposals with similar aims do not fulfil one or several of the above-mentioned criteria, revealing that the challenge remains open.
This paper presents our design science endeavour [12] from the problem investigation to the solution design, the implementation of a modelling tool, and the solution evaluation. We have chosen to integrate the languages proposed by i⁎ [3] (a goal-oriented modelling method), and Communication Analysis (CA) [13] (a communication-oriented business process modelling method). The reason for choosing i⁎ is its expressiveness to specify dependencies, with which we intend to trace strategic motivations and processes. In the case of CA, we aim to get the most out of the communicational techniques in order to analyse business processes; its notation is not what is important, but rather the underlying concepts and modularity guidelines. Moreover, some current business process modelling suites use BPMN with a communicative approach. In addition, the authors are competent in these languages and are able to confront the challenge.
The contributions of the paper are the following:
- •
We present the iStar2ca guidelines v2.0 which are intended for a top-down modelling scenario and are an evolution of the iStar2ca guidelines v1.0 reported in [14] (see the evolution time-line in Fig. 1). The guidelines design is a method engineering effort; throughout the paper, we use the terminology introduced in [15].
- •
We report a comparative experiment and a focus-group session, which was carried out with students and whose feedback has been taken into account to produce the iStar2ca guidelines V2.0. The results demonstrate that the subjects׳ effectiveness (specifically, business process model completeness) is greatly improved by the use of the iStar2ca guidelines, without compromising efficiency. Also the perceptions of the usability of the guidelines are positive.
The paper is structured as follows. Section 2 compares related works. In Section 3, we introduce and exemplify the methods selected for integration. Section 4 describes the research methodology and Section 5 details the process and proposal for integrating i⁎ and CA. Section 6 presents guidelines for a top-down modelling scenario. Section 7 describes the modelling tool and technical support. Section 8 presents how the top-down scenario guidelines were evaluated through a lab-demo, a comparative experiment, and a focus group session with students. Finally, 9 Discussion, 10 Conclusions and future lines of work conclude with a discussion and future lines of work.
Section snippets
Related work
The combination of different methods and models to obtain a profund and comprehensive understanding of the system to be produced has received much of attention in the IS literature. On the one hand, some general frameworks have been proposed to reconcile different perspectives; for instance, Salay et al. [16] propose the concept of macromodel for the development, comprehension, consistency management, and evolution of a collection of related models. On the other hand, different approaches focus
Background and running example
This section presents an overview of the languages being integrated, i⁎ and CA. Throughout the paper, we use a running example based on the case of SuperStationery Co., which is an intermediary company that buys and sells office material. The focus of the example is on the intentional and operational aspects of sales management.
Research methodology
We structure our research in terms of design science since it involves creating new artefacts and acquiring new knowledge (see Fig. 4). Our research methodology follows the cycles described by Wieringa [12]. We performed two engineering cycles (EC1 and EC2) that mainly create an integrated metamodel, the iStar2ca guidelines V0.5 and V1.0 (see Fig. 1). For the validation of the guidelines we performed a research cycle (RC1) that consists of a comparative experiment and a focus group session that
Modelling language integration
In order to combine the process and intentional perspectives, we undertake a method engineering effort [15], where i⁎ and CA are considered method chunks. Note that the analysis of project situations is out of the scope of this paper. Instead, the focus placed on integrating the product and the process models of the methods. Taking the integration map proposed in [15] as reference method, Fig. 5 presents how we have operationalised each of its intentions and points to the corresponding section.
Top-down scenario guidelines: iStar2ca guidelines 2.0
We present the iStar2ca guidelines version 2.0 for a top-down scenario, an evolution of the iStar2ca guidelines V1.0 motivated by the empirical results described in Section 8. The guidelines facilitate obtaining a CA model having as input a given i⁎ model and establishing the mappings identified at a metamodel level. They indicate how to derive ca.communicative events and ca.message structures since these CA elements only map into i⁎ elements under specific conditions. As seen in the previous
Tool support
Technological support for the GoBIS framework is necessary to carry out future case studies and facilitate industrial adoption. Although existing tools allow creating separate i⁎ and CA models, we aim to support the combined modelling (see Fig. 8 with the combined modelling for the SuperStationery Co; the CA model in the right and the i⁎ model in the left). The CE SALE 1 is aligned with the task PLACE ORDER.
We chose Eclipse (http://www.eclipse.org) as the technological platform. We used Eclipse
Validation by means of a controlled experiment
We have performed a comparative experiment to assesses the performance and perceptions of students applying the iStar2ca guidelines V1.0 [14]. This version includes Guidelines 1–7 (see Section 6). This experiment has been designed according to Wholin et al. [36], and it is reported according to Jedlitschka and Pfahl [37] and Juristo and Moreno [38].
Discussion
In this section, we present a discussion that is based on the results of the experimental task and the focus group session. Also, we discuss additional observations that are related to the complete framework presented in the paper that are worth discussing.
Conclusions and future lines of work
Given the existence of complementary perspectives in information system analysis, this work confronts the challenge of integrating a goal-based and a business process-based modelling language. A review of related works reveals that there is room for improvement. As a result, in this paper, we have reported a design science endeavour that undertakes the integration of i⁎ and Communication Analysis (CA). We have selected these languages for their expressiveness and their associated elicitation
Acknowledgements
This work has been supported by the Spanish MICINN Project ProS-Req (TIN2010-19130-C02-01, TIN2010-19130-C02-02) and EOSSAC (TIN2013-44641-P); the Generalitat Valenciana Project IDEO (PROMETEOII/2014/039); the FPI-UPV Pre-Doctoral Grant; the European Commission Project CaaS (FP7 611351); and the ERDF Structural Funds.
References (43)
- et al.
Goal-driven business process analysis application in electricity deregulation
Inf. Syst.
(1999) - et al.
Reengineering the Corporation: A Manifesto for Business Revolution (Revised)
(2003) - et al.
Strategic alignment: leveraging information technology for transforming organizations
IBM Syst. J.
(1999) Modelling Strategic Relationships for Process Reengineering
(1995)- I. Zikra, J. Stirna, J. Zdravkovic, Bringing enterprise modeling closer to model-driven development, in: P.Johannesson,...
- E., Yu and J. Mylopoulos, From E-R to “A-R” – modelling strategic actor relationships for business process...
- R.S.S., Guizzardi, et al., Bridging the gap between goals, agents and business processes, in: Proceedings of the Fourth...
- et al.
Aligning goal analysis and business process modelling: a case study in health care
Int. J. Bus. Process Integr. Manag.
(2011) - et al.
Goal-based business process models: creation and evaluation
Bus. Process Manag. J.
(1997) - G., Koliadis and A. Ghose, Relating business process models to goal-oriented requirements models in KAOS, in Advances...
Communication Analysis: A Requirements Engineering Method for Information Systems, (CAiSE׳09)
Integrating the Goal and Business Process Perspectives in Information System Analysis, (CAiSE 2014)
Using Macromodels to Manage Collections of Related Models, (CAiSE)
Towards a formal definition of goal-oriented business process patterns
Bus. Process Manag. J.
A method for eliciting goals for business process models based on non-functional requirements catalogues
IJISMD
On the notion of soft-goals in business process modeling
Bus. Process Manag. J.
Requirements-Driven Design and Configuration Management of Business Processes, BPM׳07
Cited by (18)
Including business strategy in model-driven methods: an experiment
2023, Requirements EngineeringUsing Team Topologies in Model-Driven Strategic Alignment
2023, CIbSE 2023 - XXVI Ibero-American Conference on Software EngineeringThe LiteStrat Modelling Method: Towards the Alignment of Strategy and Code
2022, Domain-Specific Conceptual Modeling: Concepts, Methods and ADOxx ToolsStra2Bis: A Model-Driven Method for Aligning Business Strategy and Business Processes
2022, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)