Executable business process modeling as a tool for increasing the understanding of business processes in an organization

(1)

!

!!!!!!!!! !

!

!!

!

Executable*business*process*

modeling*as*a*tool*for*increasing*

the*understanding*of*business*

processes*in*an*organization

! by!

Ersin!Demir!

!

Department!of!Computer!and!Systems!Sciences!

Royal!Institute!of!Technology!(KTH)!/!Stockholm!University!(SU)!

Degree!project!30!HE!credits!

Degree!subject:!Computer!and!Systems!Sciences!Degree!project!at!the!Master!level!

Spring!2014!

Advisor:!Ilia!Bider!

Reviewer:!Name!of!Reviewer!

(2)

Executable business process modeling as a tool for increasing the understanding of business processes in an organization

Ersin Demir

Abstract

Understanding of business processes is becoming an important key factor for successful businesses and today many organizations are facing the lack of knowledge about the business processes that they are working on. Since the interaction between different business processes and different actors are becoming more common it is not only enough for employees to have knowledge about the business processes that they involve directly, but also they need to know about the other business processes in the organization.

Unfortunately there is not enough research on this topic in the literature and the goal of this thesis is to propose a method for solving the indicated problem by increasing the understanding of business processes by the employees in an organization. The proposed method basically contains the design and execution of process models based on a real scenario. A case study has been conducted at an IT company where the employees have no or limited knowledge about the business processes that their organization works on. Even though the method has been only tested in one organization it is generic and can be applied to any other similar organization.

The design science approach is used to develop the method and build the process models as artifacts. The process models have been verified by using an executable business modeling tool, iPB, and presented to the employees in a seminar in order to make them to understand the business processes better. The knowledge of the employees about the business processes has been analyzed before and after the presentation, thus we could compare the results and find out how much their knowledge has increased after the presentation. The results have shown that the knowledge of the employees has increased significantly. In conclusion, the method of design and presentation of executable business process models has been proved to be a solution to the problem of not understanding of business processes in an organization well enough.

Keywords

Executable Business Process Modeling, Design Science, iPB, Artifact

(3)

Contents

1 Introduction . . . 6

1.1 Background . . . . 6

1.2 Problem . . . . 7

1.3 Goal . . . . 7

1.4 Thesis Disposition . . . . 9

1.5 Social and Ethics . . . . 9

2 Extended Background . . . 10

2.1 Business Process Management (BPM) . . . . 10

2.2 Business Process Modeling . . . . 12

2.2.1 Process Models . . . . 12

2.3 Process Modeling Tools . . . . 15

2.4 The Company . . . . 16

3 Research Methodology . . . 17

3.1 Method Overview . . . . 17

3.2 Deciding the Research Type . . . . 17

3.3 Design Science . . . . 17

3.4 Case Study . . . . 19

4 Explicate Problem . . . 20

5 Outline Artifact and Define Requirements . . . 21

5.1 Outline Artifact . . . . 21

5.2 Requirements . . . . 21

6 Design and Develop Artifact . . . 23

6.1 Analyzing the Company’s Business Process(es) . . . . 24

6.2 Choosing the Modeling Approach . . . . 24

6.3 Choosing the Modeling Tool . . . . 24

6.4 Developing the Models . . . . 24

6.5 Verifying the Models . . . . 24

6.6 Presenting the Models to the Audiences . . . . 24

7 Demonstration . . . 25

7.1 Analyzing the Case Company . . . . 25

7.2 Choosing Two Interrelated Business Processes . . . . 27

7.3 Choosing the Modeling Approach . . . . 27

7.4 Choosing the Modeling Tool . . . . 28

7.5 Developing the Models . . . . 30

7.5.1 The Sales Process . . . . 30

7.5.2 The Supply Process . . . . 32

7.6 Verifying the Models . . . . 34

7.7 Presenting the Models to the Audiences . . . . 36

8 Evaluation . . . 37

9 Discussion . . . 51

9.1 Conclusion . . . . 51

9.2 Limitations . . . . 51

9.3 Future Works . . . . 52

(4)

Bibliography . . . 53 Appendix A . . . 56 Appendix B . . . 70

(5)

List of Figures

2.1 Business Process Management Life-Cycle [37] . . . 12

2.2 IDEFØ Box and Arrow Graphics [14] . . . 13

2.3 A Workflow (Petri Net) Example [45] . . . 14

2.4 An Agent-Related (RAD) Example [38] . . . 14

2.5 A State Flow Example [41] . . . 15

6.1 Activity Framework . . . 23

7.1 Activity Framework . . . 25

7.2 Process Development with iPB [3] . . . 29

7.3 The Model of the Sales Process . . . 30

7.4 The Model of the Supply Process . . . 32

7.5 The Verification of the Sales Process . . . 35

7.6 The Verification of the Supply Process . . . 35

8.1 Survey Question 1: Sales . . . 37

8.2 Survey Question 1: Supply . . . 38

8.23 Survey Question 12: Sales and Supply . . . 48

(6)

1. Introduction

This chapter contains the background of the thesis and addresses problem, goal and research question, thesis disposition and finally describes social and ethics [17].

1.1 Background

In today’s competitive world, the organizations have become more complex in terms of business, organizational structure and IT utilization in order to be permanent in the area they exist. According to my professional experiences working at several IT companies and also my observations in IT field, this complexity causes some problems in the organizations. One of the most common problems is the inadequate understanding of personnel about the business processes that they involve and also to the other related business processes in the organization. Today especially in the big organizations this issue becomes more problematic and impacts the business efficiency. It is not enough to be only aware of the processes that the person involves directly. It is becoming more important to have knowledge about the other processes that the organization has, especially the ones that are related with individual’s work. Today in many IT companies technical personnel are rarely aware of the non-technical processes; for example an IT engineer should know how a sales process works in general perspective. The same case is available for non-technical personnel as well when it comes to know the technical processes in general. The current BPM methods mainly rely on experts too much and this causes non-experts actors to perform nothing or just few things but be affected by the process much [30].

The understanding of a business process model is relevant to three factors; personal, model and content [23]. These factors plays decisive role for either a failed or an improved understanding of the associated process models. It is mainly not easy to combine these three factors in a well defined way to improve the understanding of the models. The study done by et al Alzubidi [26] shows that the lack of process modeling understanding makes the people to get low perception of process modeling usefulness. Low level of knowledge about the business processes in an organization causes employees not to understand the benefits of process models properly. It is indicated in the same study that professional training can significantly be helpful for the employees to improve their knowledge of process modeling and this might also facilitate a better understanding within the organization about the benefits of process modeling.

It is a difficult and complex task to design a proper business process since organizational goals, business services, IT systems and value network need to be aligned. For this reason it is important to use a certain method to support the design of the business processes and also to support the evaluation of the designed processes [28]. Choosing the right modeling approach has direct relation with the success of the model and a successful model increases the understanding of the business process rapidly [4]. There many traditional approaches to business process modeling and these approaches are based on activity flows or documents. Activity flow based approach considers data as an afterthought and document based approach does the same for processing part. An emerging approach that is based on artifacts simply combines data and process in a holistic manner and it is called data-centric process modeling [7].

This thesis study has been carried out with an IT company called Synapse Mobile Networks that operates in telecommunication industry. The reasons for choosing this

(7)

company are firstly it is a small organization and there are no defined process models which will be helpful to find out easily the increase of knowledge after presenting the designed process models and secondly I work at this company and it will be easier for me to access data and collect reviews from my colleagues. Today the company faces the consequences of not having defined business process models. Most of the employees have limited knowledge about the processes that they involve directly as well as the other business processes that are used in the organization.

Synapse Mobile Networks is a small size company that has 5 different units; they are development, supply and support, sales, finance, and device test. The thesis study has been carried out with supply and support, and sales units. The reason of selecting these two units is to design a business process (supply process) that has mainly technical actors and another process (sales persons) that has mainly non-technical actors. Presenting both process models to the employees with different technical backgrounds will help us to find out the usefulness of the models and the gained knowledge about the understanding of the selected business processes by the employees.

1.2 Problem

Today in many organization most of the employees has no or limited knowledge about the business processes that their organization work on. Even though employees have better understanding of the processes they they involve directly it is not enough for a successful business. Employees mostly need to have enough knowledge about the other processes as well to enhance the business performance. It has been observed that the company where this study has been conducted with has no any defined business processes and most of the employees has no or limited knowledge about the enterprise’s businesses. The lack of this kind of knowledge causes sometimes significant problems. It is sometimes confusing to know who the responsible person is for a specific task and this causes mainly the works not to be completed on time. For example when the sales people receive technical questions they do not always sure who should answer the questions. It is also mainly a problem for technical staff to answer sales related questions due to the lack of knowledge.

It is also a problem that current process modeling notations are so complex to be used to increase the understanding of employees about the business processes that their organization has. It is even more difficult to make non-technical people to understand the models designed in a complex way [30].

This thesis focuses on to develop a process model that can be helpful for all employees from different departments to understand better the business processes that their organization work on. Unfortunately there is not enough research on this problem in the literature and with this study we aim to fill this gap and develop a method as solution to the indicated problem.

1.3 Goal

The most natural way of describing an enterprise’s business is through creating models of processes that are used by the enterprise. Well defined process models can help to improve the process efficiency. Business process modeling is mainly used for illustrating a complete process, enabling the staff to improve the business flow and streamline the process [6]. In principle it can be also used for increasing the knowledge about the processes.

However, there are some obstacles for using process modeling for this pupose such as;

(8)

• Models are usually drawn with the help of relatively complex technical understanding which is not always possible for all staff from different backgrounds to understand them easily.

• It takes to much time to understand the model if an employee does not involve in the process directly.

The overall goal of this thesis is to improve the understanding of business processes by the employees in an organization by performing a real scenario in an executable process modeling tool. Thus the models of business processes that the organization work on need to be build on a modeling tool that allows to execute the models. The execution of the models are vital for achieving the goal since it allows the employees to comprehend the models based on a real scenario. The goal also can be formulated as a research question as follows;

"How to develop and test in practice a method of business process modeling that facilitates better understanding of the processes in an organization by the employees".

Designing a proper business process model is a difficult task since different actors with different backgrounds take role in the same processes quite often. For this reason it is very important to choose the right modeling approach and tool to build the business processes. This will make the understanding of the models easier. In the scope of this thesis the built models will be presented to both technical and non-technical employees in the organization in order to understand how comprehensible the same models are to the personnel with different backgrounds.

In this thesis we will propose a method of modeling that will facilitate the understanding of the business processes models by the most employees in the organization. Further- more we will test the proposed method in practice at a company with a case study. The testing of the method will be done by following the listed steps below.

• How to choose the right modeling approach to design business processes at an IT company? In order to answer this question first of all two interrelated business processes will be chosen at the company and then secondly according to the process requirements the most appropriate modeling approach and tool will be used to build them by studying and comparing different process modeling techniques and tools in the literature.

• How useful can a process model be presented? The models of two selected processes will be presented to the participants and their feedback will help us to understand the usefulness of the models. The models will be executed in the used modeling tool to increase the understanding.

• How much are the employees aware of the processes in the organization today and how much knowledge will they gain after the study? The knowledge that the employees have about the business processes at the company will be found out after collecting data from them by using questionnaire survey and then this data will be compared to the survey data that will be collected after the presentation of the models.

(9)

1.4 Thesis Disposition

This thesis report has nine main chapters that aim to explain all the work that have been performed within this thesis. Below a brief description of each chapter is given.

Chapter 1 - Introduction: This chapter aims to introduce an overall view of this thesis study. A short background of this study is introduced followed by the existing problem, the goal and the thesis disposition. Finally the social and ethics that are taken into consideration within this thesis is described briefly.

Chapter 2 - Extended Background: It presents the literature review of the following topics; business process management, business process modeling in general and state-flow process modeling in particular, business process tools and a brief description of the company where the case study in this thesis has been conducted.

Chapter 3 - Research Methodology: This chapter describes the research methodologies in general and design science paradigm in particular since it has been followed in this thesis.

Chapter 4 - Explicate Problem: This chapter formulates the initial problem that is described in Chapter 1 in details. It explains significance of the problem and possibly analyzes its underlying causes.

Chapter 5 - Outline Artifact and Define Requirements: This chapter outlines a solution to the explicated problem in the form of an artifact and identifies the requirements for this artifact.

Chapter 6 - Design and Develop Artifact: This chapter describes both the developed artifact and process followed in the development.

Chapter 7 - Demonstration: This chapter describes how the developed artifact can be used in real world with a case study. The steps in the activity framework in Figure 7.1 are explained in detail by implementing each step in the selected organization.

Chapter 8 - Evaluation: This chapter aims to evaluate the developed artifact, ad- dressing both the defined requirements and the explicated problem.

Chapter 9 - Discussion: The last chapter of the thesis summarizes all the works that have been performed during the study. It presents also the future works that can be studied.

1.5 Social and Ethics

Since this thesis has been conducted at a private IT company then some ethical values are taken into consideration during and after the study. All the data used in this thesis has been collected information within the company and they have been informed that this study will be published publicly. During the execution of the models fake names are used in order not to reveal the private information of the employees. Another ethical consideration is to use nickname for the employees who have participated to the interviews and questionnaire surveys within this study. This is not only to protect the employees privacy but also make them to feel free when they answer the questionnaire surveys. Further- more, all scientific research ethics are taken into consideration within the thesis. All the information taken from other resources are indicated with their references completely.

(10)

2. Extended Background

This chapter presents the literature review of the following topics; business process management, business process modeling in general and state-flow process modeling in particular, business process tools and a brief description of the company where the case study in this thesis has been conducted.

2.1 Business Process Management (BPM)

Along with the recent development in information technologies the organizations have become more complex to be managed. This complexity makes Business Process Management (BPM) to be more significant and necessary for all kind of organizations.

BPM is the theory of focusing the enterprises’ business processes instead of focusing on functions [32]. Today it is not only enough to use an IT technology within the organization. The alignment of IT, organizational structure and businesses plays a vital role in order to have a successful and permanent business in today’s competitive business world. In order to minimize the complexity of a business and make it easily manageable the business processes must be well-defined and well-implemented within the organization. A business process can be defined as "a set of partially ordered activities that transform a state of something to another state with the aim of reaching certain goals" [29]. The certain goal can be to provide a service or to produce a product.

Many other definitions of business process are available, for example Devenport and Short [10] define business process as "a structured, measured set of activities designed to produce a specified output for a particular customer or market" and they put emphasis on the following three dimensions.

• Entities: The existing processes between organizational entities.

• Objects: Physical or virtual objects that are used by processes as input and/or output.

• Activities: Managerial or operational activities that contain the processes.

A business process is a map that is used to navigate through business activities. This map helps us to understand where we are now, how and from where we have come and where we should go as a next step [4]. The importance of business processes and BPM has being recognized by most organizations in the last years together with the high com- petition on cost, quality, service and time [31].

In order to have a successful business and be permanent in the competitive business environment the business processes must be well defined and well-understood by the all participants in the process [20]. Comprehension of a process is relevant how the process is modeled. Business process models are developed for improving organizational efficiency and quality. The main purpose of business process modeling is to enhance business performance by connecting activities efficiently in the supply of a product or service [6]. In order to fulfill this purpose, the business process models must be easy to understand. One of the most essential parts of a business process modeling is to choose the right modeling approach and right tools to implement the models [4]. The selection of a right approach simply increases the possibility of defining and designing a good model. As a result of a well-defined model, the process becomes easy to be comprehended. Bider [4] also states

(11)

that the adequacy of the model to the process is most essential factor in order to have a well-defined process model.

All business processes contain a number of participants which can be classified into artifacts. The notion of artifact here represents any physical or abstract object (information objects) such as documents, products, computer programs, human beings, etc. The participants can be categorized into two types: passive participants and active participants.

Passive participants are the participants that are changed during the process execution such as a document being written or a product being assembled. Active participants are the ones who perform the actions to change the passive participants. People, organizations, machines, computers, etc can be listed as active participants [5].

Hammer et al [12] states that corporations must have organizational structures and business processes that are fast, flexible, low cost and deliver high quality consistently in order to succeed in today’s global economy. The way of business management has evolved rapidly in the last decade and for traditional businesses it is unlikely to deliver the requirements that are listed above. The organizations must adapt themselves to the changing business world.

Business Process Management (BPM) is a management theory that has been developed quickly and become more significant for enterprises recently. BPM is the theory of focusing the enterprises’ business processes instead of focusing on functions [32]. Today many different enterprise solutions are available in the market which are developed based on enterprise’s requirements but the biggest difference between those all solutions and BPM is that the BPM is developed to focus on the enterprises’ processes and supports the way of managing the processes instaed of supporting the specific processes within the enterprises [35]. The BPM aims to align an enterprises’ business processes with the customer/client requirements and demands by improving them and as a consequence of this, the business efficiency and customer satisfaction are increased [43]. Another definition of BPM made by Aalst et al [44] is "BPM is used to support business processes using methods, techniques, and software to design, enact, control, and analyze operational processes involving humans, organizations, applications, documents and other sources of information". All of these definitions show us that BPM is simply a methodology to improve the business processes.

The BPM is considered as a life-cycle process and contains the following activities [16].

• Design : In this phase, existing and proposed processes are identified and re- viewed.

• Modeling : In this phaes, how the proposed design functions under various scenarios is investigated.

• Automation (Execution) : BPM includes tools that allow services and line of business applications to be connected to automate the process.

• Monitoring : Real, near real time, or ad-hoc monitoring of activities and key performance indicators within a business process enables management to react to service interruptions, bottlenecks and other inefficiencies, as well as potential risks and other threats as they occur.

• Optimization : Statistics generated by the modeling phase and process monitoring can be analyzed to identify inefficiencies and potential improvements which can be applied to the process design. Optimization contains run-time modification of processes.

(12)

As being a life-cycle process, the BPM provides the continues business process improvement. Usually changes in an enterprise’s business need to change the process models and this modification can be implemented into the process models quickly at the optimization step. Figure 2.1 shows the continues relation between the listed activities above.

Figure 2.1: Business Process Management Life-Cycle [37]

2.2 Business Process Modeling

Perjons [28] defines Business Process Modeling, also called Enterprise Modeling, as "the process of using models to represent an enterprise’s business processes, resources, people, information, constraints, and organizational structure". Business Process Modeling plays an essential role for business process improvement (optimization) that is one of the steps in BPM Life-cycle. For this reason process models becomes significant for the implementation of any process improvement projects [31]. Gong et al [31] states that process models should provide functions for the following fields.

• Description : The process representation is a form of business knowledge and the description of the model should provide a better understanding and communication of people.

• Analyze : In order to increase the process efficiency, process models should also provide some methods to be used for analyzing and supporting the process design/re-design.

2.2.1 Process Models

Today many business process modeling approaches are available in the market and they present the processes graphically generally based on process elements, such as activity and entity, with the help of time series and logical relationship. Mostly the popular modeling methods, such as IDEF3, Petri Net, EPC, RAD, UML, represent the business processes in the same way. They all at least have two basic components: a box and an arrow. The boxes are used to describe activities and entities while the arrows are used to describe interfaces [31]. In this thesis, as indicated before, it is aimed to develop the models of business processes and execute them based on a real scenario. Therefore process modeling plays a crucial role for this thesis.

(13)

Hruby [13] defines two types of business processes: exchange business processes and conversion business processes. In exchange business processes the resources are transferred by actors between each other bidirectionally. For example, when a customer buys a car, then the transferred resources between seller and customer will be money and the car. In conversion business process, the actors use or consume the resources in order to produce another resources and the produced resources can be input for other actors or they can be used in in en exchange business process. For example a car manufacturer can consume different car pieces and use machines and labor force to produce cars. Based on this, the selected business processes (sales and supply) can be defined as exchange business processes. In the sales process the company sells its IT products to its customers and in change the company gets money from the customers. As same in the supply process, the company sells its professional services in exchange of money. The supply process contains the installation and integration of the IT products on site that are sold in sales process.

There are many different parameters that can be used to measure the quality of the developed model such as precision, level of details, formal correctness, etc [5]. Bider and Perjons [5] state that "from a practical point of view, the most important parameter that describes the quality of a particular process model is its adequacy to the task for which the model will be used". The selection of the right modeling technique for the process can considerably increase the quality of the developed models. On the other hand using an unsuitable modeling approach for the given process can easily affect the quality and success of the deigned models. Ilia and Perjons [5] categorize the existing process modeling techniques in the following way:

• Input/Output flow : The process modeling approaches in this category focus on resources that are being changed, consumed, used or produced by the activities.

Passive participants are taken into consideration mainly within this category. A diagram (graph) represents the flows graphically and the activities are shown as nodes in the flows. The arrows are used to connect the activities in order to show how the output of an activity is used as input by the subsequent activities. The input/output flow does not consider the order of activities directly but it shows the casual order of activities due to the output of an activity is used by other activities as input in this category. Therefore some activities must wait some other activities to complete their tasks and produce outputs. IDEFØ is the most known approach in this category and a sample model is shown in Figure 2.2.

Figure 2.2: IDEFØ Box and Arrow Graphics [14]

• Workflow : In this category, the focus of process modeling techniques is the order of activities in time. The flows are represented by using diagrams in graphical notation. As in input/output flow, here also the nodes represent the activities while

(14)

the arrows represent the flows in a definite order. Each activity starts just after the precedent activities and ends just before the subsequent activities. There are plenty of workflow approaches available in the market, The most common ones are IDEF3 process flow diagrams, Activity Diagrams of UML and Petri Nets. A sample model of Petri Nets is shown in Figure 2.3.

Figure 2.3: A Workflow (Petri Net) Example [45]

• Agent-related view : Agent-related technique focuses on the order of the activities which are performed by agents as a part of their works. Here the agent term is used for active participants. The modeling techniques in this category describe the resources that are exchanged between agents. The resources can be information or messages. Some agent-related modeling techniques can be listed as; Role Activity Diagram (RAD), Specification and Description Language (SDL), and Business Model Language (BML). A sample model of RAD is shown in Figure 2.4.

Figure 2.4: An Agent-Related (RAD) Example [38]

• State flow : In this category the focus is on the order of states in a process. The state term can be described as the change in the part of the real world that em- braces the processes. The state changes are produced by activities. Some changes can be related with the state of passive participants such as their form, shape or physical location, while some changes can be related with the state of active participants such as the state of the mind of a human agent. The state flow approach is used in IDEF3 as a complementary view to the activity flow. Besides this, the state flow approach is used in UML State-Machine Diagram as the main focus.

The state-flow modeling technique is useful for modeling the business processes which are difficult to create an order of activities. This technique requires no pre-defined activity flow and for this reason it is appropriate for designing more flexible business processes. A sample of state flow is shown in Figure 2.5.

Within this thesis, we are going to model business processes in an organization and present them to the employees with the expectation of increasing their understanding

(15)

Figure 2.5: A State Flow Example [41]

of the business processes. In the case study we are going to model sales and supply processes at an IT company. The chosen business processes are more appropriate to be designed based on state flow modeling approach since the states of both active and passive participants are changed often during the process. The state flow technique focuses on to give the process participants the idea of in which processes they participate. The state flow approach aims to show the employees which changes are done by which user in a process state during its life cycle [41]. Since the goal of this thesis study is to increase the knowledge of the employees about the business processes that they work on then we think that the state flow modeling approach suits very well for the models that we aim to develop in this thesis. The processes are described and modeled in detail in Chapter 7.

2.3 Process Modeling Tools

The right selection of the modeling tool improves the better understanding of the models significantly. Therefore a literature review has been conducted in order to study and compare the different tools and choose the right one for building the process models.

There is no a universal single modeling tool that can be used to create all types of process models. The requirements of a process model can vary a lot. For this reason the right modeling tool should be chosen for a successful implementation of a process model.

The modeling tool should provide a collaborative environment that can be easily used to create and validate ideas, detects errors early, and transfer those ideas/designs to the other people in an understandable way [21].

Business Process Management allows organizations to design, implement and control their business strategies through process models. The benefits of such kind of implementation becomes limited if the process models are not executed or executed unsuccessfully.

The business processes that exist in an organization can not be well understood if the employees can not participate in process execution [15]. For this reason the executable modeling tools are more beneficial comparing to the non-executable tools. Because exe-

(16)

cutable tools allows users to execute different scenarios on the tool and this simplifies the understanding of the developed process models. Since the goal of this thesis is to improve the knowledge of employees in an organization about the business processes that the organization has then it is more suitable to use executable tool in this thesis to build the process models. The models will be presented to both technical and non-technical staff, therefore it is specially important for non-technical staff to comprehend the whole picture and this can be easily done through execution of the developed models based on a real scenario with using real actors and data.

There are several executable business process modeling tools that could be used in this thesis to build and execute the models. Some of the executable business modeling tools can be listed as iPB, Bizagi BPM Suite, Agiles BPMS & ECM, TIBCO ActiveMatrix, SYDLE SEED Community and Bonita BPM. Most of them requires official licenses that can not be afforded in the scope of this thesis. iPB tool has been chosen and used in this thesis. The reason of choosing iPB tool is that it provides full execution of the process models through its powerful web based GUI [3]. It has been provided by IbisSoft for free for this study.

2.4 The Company

This thesis study has been carried out with an IT company called Synapse Mobile Net- works that operates in telecommunication industry. The company is considered to be a small size company with 25 employees who work at 5 different units.

The company was founded in 2003 and provides different advanced mobile services to 3GSM operators worldwide. The main product of the company is Automatic Device Management System (ADMS) that is used more than 200 operators. The ADMS is a service enabling solution for automatic over-the-air (OTA) device configuration of services such as GPRS, MMS, email, SyncML, sync, SIP, Wireless Village, Push-to-talk, Book- marks, WAP browser, WiFi, Internet, firmware and application download services, lock and wipe services, Video Sharing etc. The system contains the complete service management process including subscriber detection, subscriber service provisioning, sending settings to subscribers, SIM Card management, Enterprise management, marketing tool and advanced business intelligence [40]. Apart from the ADMS the company delivers solutions such as Equipment Identity Register (EIR), Roaming Switching Center (RSC), Personal Roaming Center (PRC) and Dynamic Number Allocation (DNA).

Synapse Mobile Networks is market leader in Automatic Device Management and has strong partnership with Ericsson AB that is one of the largest companies in the telecommunication area [25].

(17)

3. Research Methodology

This chapter describes the research methodologies in general and design science paradigm in particular since it has been followed in this thesis [17].

3.1 Method Overview

A research can be defined as a scientific and systematic search for related information on a specific topic. Shortly, research is an art of scientific investigation. A research study aims to find an answer to a question using a research methodology. Therefore research methodology can be defined as the way to systematically solve the research question [19].

There are different research methodologies that can be used in a study. The vital point is to choose the right methodology since it has huge impact on the success of the study.

Empirical and design science methodologies are the two main approaches for research studies. Empirical research methods can be described as a class of research methods in which empirical observations or data are collected in order to answer particular research questions or solve a problem. In empirical research, researcher comes up with a theory in the beginning and then he/she develops the theory to explain what happens in the real world. The researcher aims to test the defined theory and possibly refine it in the empirical methodology [24]. The other main research methodology, design science, can be defined as "design science seeks to extend the boundaries of human and organizational capabilities by creating new and innovative artifacts intended to solve identified organizational problems" [2]. Denning et al [11] and Tsichritzis et al [9] also state that "the design science paradigm aims to solve problems by creating innovations (artifacts) that define the ideas, practices, technical capabilities, and products through which the analysis, design, implementation, management, and use of information systems can be effectively and efficiently accomplished". Rather than explaining the existing realities, the designed artifacts should create new realities and evaluate the aimed goals within these new realities [28].

3.2 Deciding the Research Type

This thesis study aims to solve an existing problem which is to increase the knowledge of the employees in an organization about the business processes that exist. It is more trying to solve an apparent problem rather then verifying a prior theory. Therefore design science paradigm has been applied in order to accomplish the defined problems in this thesis. I believe that Design Science methodology will make a strong contribution to this study such as creation of the process models as artifacts and evaluation of them in design science perspective.

3.3 Design Science

Today most of the organizations suffer from the lack of well-defined and easy understandable process models. The current business processes modeling methods mainly rely on experts too much and this causes non-experts actors to perform nothing or just few things but be affected by the process much [30]. For this reason proper modeling of business processes becomes significant to have a standard process model in the business and thereby make the processes easy-understandable for all employees within the organiza-

(18)

tion and improve efficiency. Design science is about creating new and innovative artifacts in order to solve existing organizational problems [2]. This thesis focuses on creating new proper business process models that can be used to increase knowledge about the business processes. For this reason I believe that design science paradigm is more suitable than other paradigms to be used in this thesis. The developed process models are considered as artifacts.

March et al [22] have classified four different types of outputs produced by design science: representational constructs, models, methods and instantiations. Representational constructs are concepts that define a basic language that can be used to characterize phe- nomena. Models are combination of the constructs in higher order to be used to describe tasks, situations, or artifacts. The models represent the state of existing or imagined states with some specific presentation. Methods can be defined as ways of performing goal- directed activities. These ways can be differ from algorithms and guidelines to rules and policies. Instantiations are the physical implementation of prototypes or IT systems intended to carry on certain tasks. Since this thesis aims to build business process models then the models type of output will be produced by design science paradigm in this thesis.

Design science is a problem solving process and Henver et al [2] proposes seven guidelines in order to create an IT artifact to be used to solve a problem. The proposed seven guidelines aim to assist researcher, reviewers, editors, and readers to understand all requirements to accomplish an effective design-science research. Those guidelines are ad- vised to be used in research projects, however the researchers, editors, and reviewers should use their own skills to determine how those guidelines should be applied within the research project.

The proposed guidelines are listed and discussed in detail as follow.

• Guideline 1 - Design as an Artifact : The design-science research in Informa- tion Systems aims to be innovative and solution oriented. It should be described effectively and should contain the relevant information about how to enable its implementation and application. There are four different types of IT artifacts that are accepted mainly. An IT artifact can be an instantiations, contruct, model or method [2]. Since business process models are aimed to be built in this thesis then the designed IT artifact will be in the type of model artifact.

• Guideline 2 - Problem Relevance : The main goal of design-science research is to develop IT solutions for unsolved and relevant business problems. Design science approaches this by creating innovative artifacts. Henver et al [2] defines problem as "the differences between a goal state and the current state of the system" and problem solving as "a search process using actions to reduce or elimi- nate the differences". The aimed goal within this thesis is how to design a process model that makes the employees to understand the enterprise’s businesses better by increasing the knowledge of business processes within the company. The thesis goal shows the problem relevance while the problem solutions will be modeling of business processing and presenting them to the people who have different technical background and make them to gain some knowledge about business processes in the company.

• Guideline 3 - Design Evaluation : Evaluation phase takes significant place in design-science research process. The quality and efficacy of the designed artifacts must be verified by executing some evaluation methods. Henver et al [2] lists five different evaluation methods such as Observational, Analytical, Experimental, Testing and Descriptive. In this thesis, firstly a questionnaire survey will be used

(19)

in order to find out the level of knowledge that the employees have about the selected business processes. Then the designed models of business processes will be presented by executing the models based on a real scenario. Then another questionnaire survey will be conducted with the audience in order to understand how much knowledge they gain. For this reason the evaluation methods will be Observational and Experimental.

• Guideline 4 - Research Contribution : Design-science research must at least provide one contribution to the following areas; the design artifact (enabling a solution to the problem), design construction knowledge (improve the knowledge for the existing foundations) and design evaluation method (new evaluation met- rics). This thesis study will provide a contribution to the foundation by increasing the knowledge about the existing business processes and make the staff to understand the company’s business better. The design artifact will make a contribution to solve the lack of business process knowledge within the organization.

• Guideline 5 - Research Rigor : In design-science researches rigorous methods are essential in both the creation and evaluation of the designed artifacts. The knowledge of behavioral theories and en empirical works are important to create and evaluate such designed artifacts [2]. Within this thesis study, the designed artifacts will be constructed based on the latest business processing methods and tools. The designed artifacts will be presented in the organization to technical and non-technical employees and the audience will evaluate the artifacts after the presentation.

• Guideline 6 - Design as a Search Process : In order to design an effective artifact a search process is required to utilize the available means to reach the aimed goal [34]. In order to create business process models as artifacts the data will be collected from the organization via information objects. Those information objects can be documents, online pages or employees. The collected data will be analyzed and the models will be designed based on the result of analysis.

• Guideline 7 - Communication of Research : The designed artifacts will be presented to technical (engineers) and non-technical (managers and sales people) in a seminar. Two questionnaire surveys will be distributed to the audiences, one before the presentation and one after the presentation, in order to collect their reviews and find out how much knowledge they gain after the presentation by comparing the survey results.

3.4 Case Study

There are many different ideas about the definition of case study. [18] A case study in research methodology can be defined as "the collection and presentation of detailed information for a particular situation, participants or small group in order to narrow down the research into a specific topic " [42]. Robert Stake [39] defines the case study as "A form of research, case study is defined by interest in individual cases, not by the methods of inquiry used". The case study is useful to test the proposed theory or model in the scientific context in the real world in order to verify its reality [33].

In this thesis a case study has been conducted at an IT company in order to test whether the proposed executable modeling approach work in real world or not.

(20)

4. Explicate Problem

This chapter formulates the initial problem that has been introduced in Chapter 1. It explains significance of the problem and possibly analyzes its underlying causes [17].

Problem explication has been done during the case study at the case company. As a part of the study a questionnaire survey (see Appendix B) has been conducted in order to find out how much the employees know about the business processes that the company work on. Also my observations as an employee at the company have been decisive in order to explicate the problem clearly.

As mentioned in Section 1.2, the main problem at the company is the lack of knowledge about the business processes that they use and involve actively. The interviews and results of the first questionnaire (see Chapter 8) survey proved that generally all employees had basic knowledge about the processes that they involve directly but they have less or no knowledge about the other processes. The sales people were more aware of their sales process comparing to the supply people being aware of their own supply process. Since the case company is very small and there is no any well-defined process model some problems occur when someone needs to know more about what are happening in the other business processes that exist in the organization. For example, as a project manager in support and supply it is always a need to know about the progress in the sales process to be able to plan the projects or works properly in advance. As a supply and support manager it is very important to know the latest status of the sales projects since she/he needs to plan the resources according to the upcoming projects. Today the only way to get information about the sales projects is to contact the sales people directly via email or phone. This is not the proper way to keep track of the projects’ status and also it is not always easy to reach sales people since they are quite busy and travel a lot. On the other side for example for a sales person it is always a need to know how the integration projects are going since it is important to propose times, resources, works or other things to the customers when they try to sell the products. According to the observations that have been done in the organization, less knowledge about the other processes sometimes causes communication problems and thus the enterprise’s business can be impacted in a negative way. This can result lose of time, effort, money and reputation.

As a result two main problems have been observed at the case company.

• The employees in general have limited knowledge about the business processes that they involve directly and have less or no knowledge about the other business processes that they involve partly.

• There is no any well defined business process at the company and this causes the employees not to understand the business processes properly.

(21)

5. Outline Artifact and Define Requirements

This chapter outlines a solution to the explicated problem in the form of an artifact and identify requirements for this artifact [17].

5.1 Outline Artifact

Design science paradigm aims to solve existing problems by creating innovations (artifacts) that define the ideas, practices, technical capabilities, and products through which the analysis, design, implementation, management, and use of information systems can be effectively and efficiently accomplished [9] [11]. There are four different types of outputs produced by design science: representational constructs, models, methods and instantiations [22]. The overall goal of this thesis, as described in Section 1.3 is to increase the understanding of employees in an organization about the business processes by developing and testing in practice a method of business process modeling. The proposed method of business process modeling (see Figure 6.1) is the main artifact in this thesis. With the developed artifact, it is aimed to create better understanding of business processes by the employees. Even though the artifact has been tested only at a small IT company in a case study, the approach can be applied to any similar size and type of IT companies that suffer from the lack of well defined business processes. The thesis focuses more on how to develop a generic artifact that can be used in many organization to increase the knowledge about the business processes rather than focusing on some specific organization with specific business processes.

In order to achieve the defined goal, the following steps need to be completed to outline the artifact.

• Choose a modeling approach: In this step, modeling approaches are analyzed and based on the requirements the right modeling approach is chosen. It is very important to choose the right modeling approach to develop the artifact. The use of any inappropriate modeling approach may cause problems for the understanding of the artifact.

• Choose a modeling tool: After choosing the modeling approach then a modeling tool is chosen. Selection of the right modeling tool is important as the modeling approach for developing a successful artifact.

• Build the models: Based on the collected data from the company, the models are built by using the chosen modeling approach and tool.

• Present the models: In the last step the models are presented to the employees in order to improve the knowledge about the business processes within the organization.

5.2 Requirements

In order to gather the requirements of the artifact, a research work has been conducted in the company. The collection of information has been done mainly through information objects, interviews and questionnaire surveys. The collected data has been analyzed ac-

(22)

cording to the theoretical study done by reviewing the related literature. There are some main criteria taken into consideration when defining the requirements. They are such as;

• Who are the relevant employees and what kind of technical background do they have?

• What kind of information is used/changed/transferred inside the organization and also with suppliers, customers and partners?

• What are the most significant services/works for the company?

Based on the analysis of collected data in the company and the criteria above, the following requirements are defined for each step that outlines the artifact as described in Section 5.1.

• Requirements for choosing a modeling approach: The artifact should be easily understandable for all employees who have different technical backgrounds. It is very important to build a process model that can be suitable for all employees in the organization. For example, in our particular case study at the case company, the models will be presented to both engineers and sales people at the same time. Having too many technical details makes the models to be complex for the sales people to understand them. The success of this thesis is dependent on how much the employees understand and increase their knowledge about the modeled business processes.

• Requirements for choosing a modeling tool: The modeling tool should suit with the modeling approach well. Otherwise it is unlikely to design successful process models. The right modeling tool facilitates the designing and understanding of the process models. The modeling tool should also allow us to execute the models easily without a complex technical procedure.

• Requirements for building the models: There is no a specific requirement for this step. In general, the models should be built properly based on the collected information during the case study by using the chosen modeling approach and tool.

The models should also contain all details of the business processes. Any missing information in the models may cause misunderstanding during the presentation of the models.

• Requirements for presenting the models: The models should be easily presented and understood by the audiences. For this reason the models should be easily executed based on a real scenario in order to make the audiences to understand better.

The listed requirements above are taken into consideration when developing the artifact. This thesis study aims to fulfill those requirements for successful artifact to reach its goal.

(23)

6. Design and Develop Artifact

This section describes both the developed artifact and the process followed in the development of the models [17].

A method of business process modeling is proposed in this thesis to build the business processes in an organization and execute them based on a real scenario with the purpose of finding a solution to the explicated problem in Chapter 4. As indicated in Chapter 3, the design science paradigm has been applied within this thesis and all proposed guidelines have been considered as the main reference during the whole study as well as in the design and development of the artifact. The main artifact in this thesis is the proposed method of business process modeling that facilitates better understanding of the processes by the employees. The developed artifact should fulfill the defined requirements listed in Chapter 5and provide an answer to the research question. The literature review has shown that there is no any research similar to this study and for this reason with the developed artifact it is aimed to fill this gap in the literature as well.

The activities, as shown in Figure 6.1 Activity Framework, have been taken as a reference of generic artifact in this thesis to design and develop the artifacts. In the first step the company is analyzed in order to understand its general organizational structure and the businesses that they do. Based on the analysis the company’s business processes are identified. The data is collected in this step to understand the requirements of the business processes in details. In the second step, available process modeling approaches are studied and based on the defined requirements in the first step the right modeling approach is chosen. Then similarly the right modeling tool is chosen to be used for the development of the business processes in the next step. This thesis focuses on usefulness of executable modeling methods for increasing the understanding of business processes. So the chosen tool must be an executable tool that should allow us to execute the processes based on a real scenario. In the verification step the models are executed with real data based on a real scenario in order to verify the accuracy of the models. If the models are not verified then we get back to previous step to re-develop the model. This cycle is performed until the models are verified. After verification step the models become ready to be presented to the employees.

Figure 6.1: Activity Framework