Available online at ScienceDirect. Procedia Engineering 182 (2017 )

Transcription

1 Available online at ScienceDirect Procedia Engineering 182 (2017 ) th International Conference on Engineering, Project, and Production Management Does Experience Matter? Factors Affecting the Understandability of the Business Process Modelling Notation Renata Gabryelczyk a, Arkadiusz Jurczuk b,* a University of Warsaw, Faculty of Economic Sciences, Dluga 44/50, Warsaw , Poland b Bialystok University of Technology, Faculty of Management, Wiejska 45A, Bialystok , Poland Abstract This study proposes and tests an instrument for measuring the understandability of the selected business process modelling notation. Based on empirical research, we evaluated differences in the understandability of business process modelling notations between the groups of respondents, experienced and inexperienced in business processes modelling, and we identified the features of notations that determine the understandability of the modelling technique. Three notations were subjected to the diagnosis: EPC, BPMS, and BPMN. The analysis of differences in the understanding of a notation may be helpful when testing the process-related competence of people participating in process modelling and analysis projects The Authors. Published by Elsevier by Elsevier Ltd. This Ltd. is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the organizing committee of EPPM2016. Peer-review under responsibility of the organizing committee of EPPM2016 Keywords: business process modelling; modelling notations; understandability; BPMS; BPMN; EPC 1. Introduction Business process modelling is a necessary and initial part of Business Process Management (BPM) and the key element in a process-driven development of organisations. Documentation and standardization of processes in the form of graphical models require the involvement of employees from different departments and management levels, with different competencies, as well as external consultants [10]. It is therefore important that all of them use the same and understandable modelling notation which is also referred to as Business Process Modelling Technique (BPMT), [7, 9, 20]. For this reason the analysis of factors affecting the understandability of the business process modelling * Corresponding author. Tel.: ; fax: address: a.jurczuk@pb.edu.pl The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the organizing committee of EPPM2016 doi: /j.proeng

2 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) notation motivates our study. The competence of understandability of the process models using adequate methods is essential to the development of process maturity of the organization [13]. There are three main aims of this paper. The first one is to propose and test a measuring instrument of understandability of selected business process modelling notations based on empirical research. The second aim is to evaluate differences in understandability of business process modelling notations between the groups of respondents, experienced and inexperienced in business processes modelling. In particular, we assume that the understandability of Business Process Modelling Techniques in respect of each individual notation analysed (Event-Driven Process Chain EPC, Business Process Management System BPMS, Business Process Model and Notation BPMN) will be higher for the group experienced in business process modelling (EXP) than for the group inexperienced in business process modelling (INEXP). The third aim is to identify the features of notations that determine the understandability of the modelling notation. Thus, the main research questions of this paper are as follows: RQ1: What is the diversity of the understandability of business process modelling notations between the groups of experienced (EXP) and inexperienced (INEXP) in business process modelling? RQ2: What features of notations determine the understandability of the modelling notations? The factors affecting the understandability of the business process modelling notation may be useful both in choosing an appropriate modelling notation and/or business process modelling software, as well as testing the processrelated competence of people participating in process modelling and analysis projects. The paper has been structured as follows: the theoretical foundation of the study is presented in the next section, followed by the presentation of research plan and methodology used in the study, which is inscribed in the conceptual framework of the paper. The interpreted and discussed results are presented in the next section. In summary, there are conclusions and future research plans, which simultaneously should overcome the limitations of the current study. 2. Background 2.1. Business process modelling: application areas Coherent and still current application fields of business process modelling distinguish two major areas: the design and improvement of an organisation, and the design and implementation of IT systems [19]. The first area includes primarily Business Process Management implementation, but also process oriented reorganisation, certification, benchmarking and knowledge management. The second field concerns IT system choice and development, model-based customization, workflow management, and simulations. Depending on the modelling perspective adopted in the organisation s project and their purpose and scope, different techniques may be appropriate [5, 11], but the key criterion for selection of BPMT should be high understandability of models. Process defining, documenting in the form of maps and models, and its understandability stabilizes the initial stage of the BPM implementation and enables the organization to move to a higher level of maturity and achieve higher efficiency [13]. Modelling using adequate techniques and tools is among fundamental services, which offer the BPM centers of excellence in organizations that consciously design their own strategies for the BPM implementation and development [20] Understandability of business process modelling techniques: related works The process model, in a pragmatic approach, is well represented by BPM technique when it supports a cognitive processes effectively in reasoning with the model [4, 6]. Understandability of BPMT, in this perspective is an extent to which the technique can be understood by its users not having specialist knowledge of the technique [9, 21]. A common measures of BPMT understandability are the perceived ease of understanding a model (subjective measure) and correct answers on the model content (objective measure) [3]. These measures describe the user s ability to read and interpret process model prepared with a certain BPM technique. An examination of understandability of EPC and BPMN notation was conducted by Jošt et al., but they investigated only intuitive understandability of diagrams, not objective, and compared additionally with UML AD notation [7]. The model comprehension and perceived ease of understanding were analysed by Recker and Dreiling concerning only two notation e. g. BPMN and EPC [17]. The influence of model characteristic on a pragmatic quality of modelling notation was published only by a few researchers. Mendling and Strembeck analysed which element of model characteristic

3 200 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) contributes to a good understandability only of EPC notation [15]. Genon et al. verified a cognitive effectiveness concerning BPMN notation [4]. Figl et al. investigated whether the symbol sets in visual process modelling languages EPC, UML AD, YAWL and BPMN influence the comprehension and cognitive load of process models regarding to individual competencies of respondents [1]. According to our best knowledge there has been hardly any work on understandability of EPC, BPMS and BPMN notations in our research perspective. 3. Research Plan and Methodology 3.1. Conceptual framework According to our research plan and methodology, we propose conceptual framework logically structured according to the literature and technical foundation, and methodology used in the study, which allowed obtaining research results. Based on literature review and using BPMT, we have chosen the original set of notations and have built a survey for examination of their understanding. We used the method of the structured survey in the form of an electronic questionnaire. Literature and technical foundation Literature review on Business Process Modelling as a part of Business Process Management Literature review on understandability of the business process modelling notations Modelling using BPMT such as: BPMS, BPMN, EPC Methodology used in the study Survey among the groups: INEXP and EXP Proposition of BPMT understandability indicator Quantitative methods: descriptive statistics, Mann- Whitney U Test, Correlation analysis Qualitative methods: perceived competence on Business Process Modelling Results of research Measuring instrument of BPMT understandability Comparative analysis of BPMT understandability between INEXP and EXP Features of notations determining the understandability of BPMT References of BPMT usage due to users competences Fig. 1. Conceptual framework Data collection and sample characteristics The data was collected from November 2015 to May 2016 through an on-line questionnaire-based survey addressed first to employees from companies and public sector institutions declaring the implementation of projects that require business process modelling. Secondly, we conducted a survey among economics and business administration students (mostly experienced groups). We collected a sample of 308 valid surveys on BPMT understandability, respectively 172 respondents (55.8%) participating in the survey declared to have had no experience at all in business process modelling, and 136 (44.2%) respondents declared to have had experience in business process modelling, which was confirmed by answers obtained to questions about the use of modelling techniques (Table 1) and questions about the objectives of business process modelling.

4 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) Table 1. Sample description (n=308). Items Sample size Experience in BPMT using (only EXP respondents) Sample composition Number of respondents Percentage INEXP % EXP % BPMS % BPMN % EPC % Other % 3.3. Business Process Modelling Techniques used in the study Our study has covered three BPM techniques based on an activity diagram and presenting a flow of the process: EPC, BPMS and BPMN. EPC has been developed in 1992 by the team of August-Wilhelm Scheer in a joint research project with SAP AG and IDS Scheer AG [8]. BPMS has been established by Dimitris Karagiannis from the University of Vienna in 1995 and is developing by BOC Information Technologies Consulting GmbH as a result of science and business cooperation [2]. BPMN has been developed in 2002 by Stephen A. White in the scope of Business Process Management Initiative (BPMI) and currently is maintained by the Object Management Group since the two organizations merged in 2005 [16]. The most central symbols have a counterpart in these modelling notations. These techniques have been used for modelling a generic process that describes Processing a freight forwarding order. The model represents a business process flow (a basic level of each BPMT). The number of features (shapes, colours) for BPMT is comparable and comes from a structure of each technique. Moreover, the number of elements in each diagram (activities, parallel splits, merges, loops and business roles) is the same for each process model. The level of complexity and depth of the process model does not require the respondents expertise in the domain knowledge. Understandability of the process s description has been positively examined and validated by authors in the pilot study [3] Understandability measurement instrument Understandability measurement instrument based on the respondents perceived (subjective) and verified (objective) understanding of the business process modelling notations. The test of notation understandability included a verification component, which enabled objective assessment of respondents understanding of the notation. The business process modelling notation understandability indicator was defined as a ratio of the number of all respondents who answered the questions verifying the subjective perception of indicated notation correctly to the number of all respondents who perceived the indicated notation as understandable. The indicator is a synthetic representation of the notation understandability, allowing a comparison of analysis results for all notations covered by the study and for both groups of respondents. Comparison analyses were performed to evaluate if two groups, INEXP and EXP, are significantly different from each other. Furthermore, the self-assessment of knowledge and skills on business process modelling, included in the survey, was used to investigate the correlation between the understandability of BPMTs and perceived competence. Selfassessment of BPM competence is a common and approved approach used in the survey of understandability [14]. To identify the features determining the choice of BPM techniques, we have analysed five features of the flowchart diagram, e.g. number, colour and shape of the graphic symbols, a graphic way of describing business roles and decision points (logical operators) that may influence an understandability of the process model prepared in the analysed BPMT.

5 202 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) Results and discussion 4.1. Understandability of the BPMT: comparative analysis Comparative analysis of BPMT understandability conducted for each notation by groups of respondents inexperienced and experienced indicates that the most understood, both subjectively and objectively is BPMS intuitive, business notation. Surprisingly, there is a lack of statistically significant differences in the understandability of BPMN and EPC advanced, technical notations (Table 2). Table 2. Differences in the understandability of BPMT between INEXP and EXP groups. BPMT INEXP EXP Perceived understandability Understandability indicator Perceived understandability Understandability indicator BPMS 91.9% 51.2% 83.1% 65.4% 0.046* BPMN 59.3% 42.1% 68.0% 47.8% 0.490* EPC 69.1% 35.0% 63.2% 33.7% 0.872* Significance level * (p<0.05). Mann Whitney U Test Results of the self-assessment of competence on business process modelling also allowed us to make a comparative analysis between EXP and INEXP. It is obvious that we are definitely observing the higher declared level of knowledge and skills on business process modelling for the group of experienced respondents. The experienced group of respondents perceives their competence at the moderate level, while the group of inexperienced as low or indicate the lack of knowledge and skills in the area of process modelling. Table 3. Perceived competence on Business Process Modelling comparison between EXP and INEXP. BPM competence INEXP EXP BPMS BPMN EPC BPMS BPMN EPC average value (median) I know benefits of using business process modelling 2.69 (3) 2.74 (3) 2.13 (2) 3.48 (4) 3.90 (4) 3.77 (4) I know and I am able to use BPMT 2.00 (2) 2.19 (2) 1.58 (1) 3.02 (3) 3.45 (4) 3.31 (3) I could explain differences between maps and models 1.97 (2) 2.03 (2) 1.65 (1) 3.00 (3) 3.24 (3) 3.35 (3.5) I could conduct comprehensive process characteristic 1.85 (2) 1.81 (1) 1.58 (1) 2.82 (3) 3.16 (3) 3.27 (3) I could define a business process architecture 1.85 (2) 1.74 (1) 1.48 (1) 2.66 (3) 3.18 (3) 3.46 (3.5) I know tools for business process modelling 1.93 (2) 1.90 (2) 1.42 (1) 2.87 (3) 3.45 (3) 3.15 (3) According to sample characteristics (Table 1) the most respondents from EXP group have experience in the use of EPC notation. Simultaneously, the respondents from EXP group assess their competence on process modelling the highest and show the lowest level of understandability of EPC notation, even lower than respondents from the INEXP group (Table 3). In order to observe if the understandability of one notation is correlated to the understandability of another technique, and to examine the importance of perceived competence in this understandability, we used correlation analysis (Table 4).

6 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) Table 4. Spearman s correlation coefficients of BPMT for INEXP and EXP groups. BPMT INEXP EXP BPMS BPMN EPC BPMS BPMN EPC BPMS BPMN * * EPC * * Significance level * (p<0.05). The correlations coefficients of all BPMT in INEXP group are very weak, but positive and significant, which means that if the group of inexperienced respondents understood one notation often understood also other BPMT. We do not observe a significant correlation between notations in EXP group with the exception of understandability interdependence between BPMS and BPMN Features of notations determining the understandability of the BPMT The high level of BPMT understandability means that a graphical representation of a process is clear and easily understood by its users. From a pragmatic point of view BPMT understandability i. a. depends on a graphical readership (graphic symbols and layout) and pattern recognition. These features build cognitive insights regarding the process of reading and understanding a process model [12, 21]. The results of identification of features which determine the readership and consequently a choice of the BPMT as the easiest-to-understand is given in Table 5. Table 5. Features determining the choice of the easiest-to-understand BPMT. Features of notation INEXP EXP Mann Whitney U Test modal value (median) Number of the graphic symbols 4 (4) 4 (4) 0.346* Shape of the graphic symbols 4 (3) 3 (3.5) 0.022* Colour of the graphic symbols 4 (3) 3 (3) 0.886* Graphic way of describing business roles 4 (4) 4 (4) 0.617* Graphic way of describing decision points 4 (4) 4 (4) 0.995* Significance level * (p<0.05). The result of the Mann Whitney U Test shows that the shape of the graphic symbols determines the choice of the easiest-to-understand notation in analysed groups (EXP, INEXP). Modal values of analyzed features of notation suggest that the shape and colour of the graphic symbols is more important for inexperienced respondents. Other features of the notation do not differentiate a decision about the choice of easiest-to-understand notation by inexperienced and experienced respondents. 5. Conclusions, limitations and future research Our paper presents a fundamental aspect of tailoring of business process modelling notation to preferences and competencies of their users. Regarding the results from our survey we can conclude that experienced users as well as inexperienced perceive an intuitive BPMS notation as the easiest to understand. This notation has the highest understandability indicator among EXP and INEXP respondents. BPM practitioners perceive also BPMN as easy-comprehensible notation. EPC has the lowest understandability indicator although the majority of respondents have gained experience in modelling in this notation (41%). Such a result may indicate a low quality of BPM projects in which experienced respondents participated. Furthermore, the shape of the graphic symbols the most strongly determines the choice of the easiest-to-understand notation in two analysed groups. The experienced respondents

7 204 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) perceive their competence at a moderate level, but they are not sufficient to fully understand the process description in the examined modelling techniques. The value added of this study is, on the one hand references of usage of an intuitive notation in BPM projects and on the other hand the evidence of existing a competence gap in business process modelling. This research also has some limitations that are mainly related to the self-assessment of business process modelling competence of respondents and the set of factors influenced the BPMT understandability [15, 17]. As a part of the future research other factors e.g. a model purpose and a problem domain [18] should be taken into account. Furthermore, the verification of BPM competence and BPMT understandability ought to be based on problem-solving experiment [6, 17]. Acknowledgements 7th International Conference on Engineering, Project, and Production Management (EPPM2016) was financed in the framework of the contract no. 712/P-DUN/2016 by the Ministry of Science and Higher Education from the funds earmarked for the public understanding of science initiatives. 7th International Conference on Engineering, Project, and Production Management (EPPM2016) finansowana w ramach umowy 712/P-DUN/2016 ze środków Ministra Nauki i Szkolnictwa Wyższego przeznaczonych na działalność upowszechniającą naukę. 7th International Conference on Engineering, Project, and Production Management (EPPM2016) was co-organised by the Agency for Restructuring and Modernisation of Agriculture (Poland). References [1] Figl K, Mendling J, Strembeck M. The influence of notational deficiencies on process model comprehension. Journal of the Association for Information Systems 2013;14(6): [2] Fill HG, Karagiannis D. On the conceptualisation of modelling methods using the ADOxx meta modelling platform. Enterprise Modelling and Information Systems Architectures-An International Journal 2013;8(1):4 25. [3] Gabryelczyk R, Jurczuk A. The diagnosis of information potential of selected business process modelling notations. Information Systems in Management 2015;4.1: [4] Genon N, Heymans P, Amyot D. Analysing the cognitive effectiveness of the BPMN 2.0 visual notation. In: Malloy B, Staab S, Van Den Brand M, editors. Proceedings of the Third International Conference on Software Language Engineering, Berlin Heidelberg: Springer-Verlag; 2011, p [5] Giaglis GM. A Taxonomy of Business Process Modeling and Information Systems Modeling Techniques. The International Journal of Flexible Manufacturing Systems 2001;13(2): [6] Hahn J, Kim J. Why are some diagrams easier to work with? Effects of diagrammatic representation on the cognitive integration process of systems analysis and design. ACM Transactions on Computer-Human Interaction 1999;6(3): [7] Jošt G, Huber J, Heričko M, Polančič G. An empirical investigation of intuitive understandability of process diagrams. Computer Standards & Interfaces 2016;48: [8] Keller G, Nüttgens M, Scheer A-W. Semantische Prozeßmodellierung auf der Grundlage Ereignisgesteuerter Prozeßketten (EPK), [Semantic process modelling based on Event-driven process chains (EPC) ], Institut für Wirtschaftsinformatik, 1992;89:1 30. [9] Kock N, Verville J, Danesh-Pajou A, Deluca D. Communication flow orientation in business process modelling and its effect on redesign success: results from a field study. Decision Support Systems 2009;46: [10] Krcmar H, Schwarzer B. Prozeßorientierte Unternehmensmodellierung Gründe, Anforderungen an Werkzeuge und Folgen für die Organisation [Process-oriented enterprise modeling reasons demands on tools and implications for the organization]. In: Scheer A-W, editor. Lehrstuhl für Wirtschaftsinformatik, Wiesbaden: Gabler Verlag; 1994, p [11] Kunze M, Luebbe A, Weidlich M, Weske M. Towards Understanding Process Modeling The Case of the BPM Academic Initiative. Business Process Model and Notation, Springer Berlin Heidelberg, 2011;95. [12] Lindland OI, Sindre G, Solvberg A. Understanding Quality in Conceptual Modeling. IEEE Software 1994;11(2). [13] Locamy A, McCormack K. Development of a chain management process maturity model. Supply Chain Management: An International Journal 2004;9(4): [14] Mendling J, Reijers HA, Cardoso J. What makes process models understandable? In: Alonso G, Dadam P, Rosemann M, editors. Business Process Management, Heidelberg: Springer; 2007, p [15] Mendling J, Strembeck M. Influence factors of understanding business process models. In: Abramowicz W, Fensel D, editors. Business information systems, Berlin: Springer Berlin Heidelberg; 2008, p [16] Object Management Group, Business Process Model and Notation (BPMN), Version 2.0; 2011.

8 Renata Gabryelczyk and Arkadiusz Jurczuk / Procedia Engineering 182 ( 2017 ) [17] Recker J, Dreiling A. The Effects of Content Presentation Format and User Characteristics on Novice Developers Understanding of Process Models. Communications of the Association for Information Systems 2001;28: [18] Reijers HA, Mendling J. A study into the factors that influence the understandability of business process models. IEEE Transactions on Systems, Man, and Cybernetics- Part A: Systems and Humans 2011;41(3): [19] Rosemann M, Schwegmann A, Delfmann P. Vorbereitung der Prozessmodellierung [Preparation of process modelling]. In: Becker J, Kugeler M, Rosemann M, editors. Prozessmanagement, Berlin Heidelberg: Springer; 2005, p [20] Rosemann M, de Bruin T, Power B. BPM Maturity. In: Jeston J, Nelis J, editors., Business Process Management: Practical Guidelines to Successful Implementations, Burlington: Elsevier; 2008, p [21] Schrepfer M, Wolf J, Mendling J, Reijers HA. The Impact of Secondary Notation on Process Model Understanding. In: Persson A, Stirna J, editors. The Practice of Enterprise Modeling, Stockholm: Springer Berlin Heidelberg; 2009, p