1
MSc. Thesis Business Administration: Organizational & Management Control
Centralizing Versus Decentralizing The Management Information System Case Study: Sub Unit Air France KLM Asia Pacific Passenger Business
Abstract:
The managers of the sub unit of Air France Asia Pacific Passenger Business are not fully satisfied with the current MIS capacity and its information processes. The databases, related programs, and sharing system do not meet the managers’ requirements based on their task circumstances: the uncertainty and equivocality. The MIS capacity depends on the level of centralization of the information processes. The MIS is successful when its capactity meets the managers’ requirements. As for this sub unit, the match is not successful according to theory and the case study. The slightly dissimilar needs of the managers of different levels are not met by the current MIS. The information processes regarding the production and usage of information should become more centralized. This suits the moderate to low uncertainty, low equivocality, and the required narrow scope of information for the regional and area managers. These circumstances do not need a divergent set of numerous tools and programs. This creates an overload of irrelevant options and information. Furthermore, the sharing system should become more decentralized to comply with the irregular demand of information by the central managers. This hybrid system solves these inequalities and aligns the information to surmount the current data incongruity between managers.
Keywords: Management Information System (MIS), Information Processes, Task Circumstances, Information Characteristics, Success Factors, MIS Centralization.
University: University of Groningen
Faculty: Faculty of Economics and Business
Degree programme: MSc. BA – Organizational & Management Control Course: EBM859A20 – MA Thes.BA O&MC
Supervisor: J. Gusc
Supervisor 2: W.J. van Elsacker Supervisor KLM: H. van Werven
Name: R. de Roy van Zuidewijn
Address: Laurierstraat 71-III, 1016PJ Amsterdam
Phone: 06 – 4161 7327
Email: roger_de_roy@hotmail.com Student number: s1482475
2 Table of contents Page Chapter 1 – Introduction 4 1.1 Introduction 4 1.2 Management Problem 5 1.3 Research Objective 6 1.4 Limitations 7 1.5 Research Question 8 1.6 Research Methodology 8 1.7 Relevance 10 1.8 Research Design 11
Chapter 2 – Literature Review 13
2.1 Introduction 13
2.2 Terminology 13
2.2.1 Management Information System (MIS) 13
2.2.2 Management Information Processes 14
2.2.3 MIS Centralization 15
2.3 MIS Structure 16
2.3.1 Introduction 16
2.3.2 Composition MIS and Information Processes 16
2.4 MIS Requirements 19
2.4.1 Introduction 19
2.4.2 Contingency MIS 19
2.4.3 Contingency Variables MIS 20
2.4.4 Information Characteristics 23
2.4.5 Contingency Variables and Information Characteristics 24
2.4.6 MIS Success Factors 25
2.4.7 Contingency Variables and MIS Success Factors 28
2.5 MIS Capacities 30
2.5.1 Introduction 30
2.5.2 Centralized MIS Capacities 31
2.5.3 Decentralized MIS Capacities 33
2.6 Theoretical Match Between MIS Requirements and Capacities 35
3
2.6.2 Contingency Variables and MIS Centralization 36 2.6.3 Information Characteristics and MIS Centralization 40
2.6.4 MIS Success Factors and MIS Centralization 40
2.6.5 Theorectical Match Requirements and Capacities 41
2.7 Conceptual Model 41
2.7.1 Introduction 41
2.7.2 Relationships Conceptual Model 42
2.7.3 Application Conceptual Model 44
Chapter 3 – Case Study 45
3.1 Introduction 45 3.2 Results 46 3.2.1 Observations 46 3.2.2 Documentation 52 3.2.3 Survey 55 3.3 Analysis 58 3.4 Discussion 66 Chapter 4 – Conclusion 71
Chapter 5 – Further Research 77
Chapter 6 - Acknowledgements 77
7. References 78
4
Chapter 1 – Introduction
1.1 Introduction
During my internship at Air France KLM, the Management Information (MI) supervisor and I discussed several issues faced by the sub unit Asia Pacific Passengers’ Business. One of the problems regards the Management Information System (MIS) and its information processes. The managers of the sub unit at different hierarchical levels face difficulties and irritations in the information processes. These processes include the production, usage, and sharing of information (Lee and Leifer, 1992). These three aspects are not optimally arranged to satisfy the managers during the execution of tasks. The current information processes demand time-consuming, overlapping, and exhausting efforts by the managers. Since the execution of tasks within the sub unit relies on the MIS and information processes, it is necessary to investigate how these are currently organized and applied by the managers and how these could be improved. To solve the issues with the MIS, the current and alternative capacities of information processes are elucidated. An essential question in the organization or operation of the MIS and its information processes is whether these should become more centralized or decentralized (Mozkowitz and Murnighan, 1975). Or would a hybrid MIS, a combination of centralized and decentralized information processes, be more satisfying and useful to the MIS users? The success of the MIS is determined by whether the managers’ needs regarding the information processes are met under certain task circumstances. Eventually, the MIS and its information processes should support the managers’tasks. Therefore, a conceptual framework needs to be developed to provide insight into the MIS and the processing of information throughout the sub unit of Air France KLM. The goal is to find a successful MIS where the MIS capacity meets the requirements of the managers. Ultimately, the parts of the MIS could become either more centralized or decentralized until the system’s capabilities match the managers’ requirements. The requirements are based on the different task and circumstances of the managers.
5
1.2 Management Problem
The MI supervisor of the sub unit of Air France KLM indicates the following dilemma or management problem: should this sub unit centralize or decentralize its MIS processes according to the managers’ task circumstances? There are many alternatives based on centralized, decentralized, and hybrid information processes. When Air France and KLM merged in 2004, the numerous divisions had to deal with these alternatives. Several departments totally synchronized or integrated both the Air France and KLM MIS into one centralized system to benefit their information processes. However, at other units this was too difficult to achieve, costly or unecessary to arrange at that time. This also applied to the Asia Pacific Passengers’ Business Unit. Ever since the amalgation, both Air France and KLM have been using their own programs and databases. Moreover, the MIS processes, capabilities, and accessbility differ per manager as well when looking at either MIS.
Nowadays, it is questioned whether this separation of the MIS and its capabilities in this sub unit is indeed the most successful solution to support the execution of tasks of managers. The managers at this unit are divided over three hierarchical levels. These levels or positions are depicted in the organogram in appendix I. The top level management of the sub unit consist of the vice president who leads this division. Underneath the vice president there are two area managers. These managers report to the vice president. Finally, there are eight regional managers based in Asia Pacific at different regions. These managers report to the area managers. In addition to the hierarchical reporting, all managers communicate and exchange information with each other. The management positions consist of employees of both Air France and KLM. The sub unit is dispersed over Amsterdam, Paris, and several cities in Asia. All the managing positions, including the vice president, area, and regional managers, entail different tasks to maintain the passengers’ business of Asia Pacific. Due to the diverse tasks the managers or MIS users have their own needs regarding the information processes. The production and using of information is arranged via diverse databases and related programs. The sharing of information is executed via a specific exchanging tool. The question is whether the current ways of information processing is sufficient or should each or one of the systems become more centralized?
6
1.3 Research Objective
This thesis tries to find a successful match between the managers’ requirements and the level of centralization of the MIS processes from the perspective of the managers’ task circumstances. The literature review addresses this issue in general, while the case study applies the theory on the sub unit Asia Pacific Passenger Business at Air France KLM.
Firstly, the managers’ requirements are based on the work context and the needed information and corresponding systems. These determinants are related to success measurements to verify whether the managers are satisfied with the MIS and what could be changed. The managers require a certain interpretation of the MIS to comply with their task circumstances. The databases, tools, programs, and sharing options should support the managers in their activities. In this case, the level of centralization of the MIS is examined since this has a great share in the MIS structure and its operation (Mozkowitz and Murnighan, 1975). In some cases, managers are better of with centralized information processes, while other circumstances require a more decentralized MIS. Success factors should confirm the right level of centralization and match between the managers’ requirements and MIS capacitities (Lee and Leifer, 1996). This set-up is chosen since it involves the managers’ surroundings and the appropriate information and level of centralization of the MIS which are eventually related to success factors. The idea is to show managers’ requirements regarding information and systems under different task circumstances. These circumstances vary alongside different variables. The literature review investigates what variables are commonly used to describe the nature of tasks. Eventually, these contextual factors are linked to information and system requirements which should be matched by MIS capacities. This match is based on both theory as well as on reality through a case study. The requirements are investigated alongside theory and applied within the case study to discover practical insights. To measure the success of the requirements and the match, success factors are involved. These focus on the quality of the information, systems, and service. The literature review discovers what factors should be involved to determine the success of the MIS. The goal of the literature review is to examine the MIS structure, determinants, level of centralization, and capacities to establish successful information processes. Moreover, this theory is applied in the case study. The collected theories lead to a conceptual model. This framework is used in a case study to elaborate on a data collection about the sub unit of Air France KLM. The goal of the case study is to discover whether the managers are satisfied with the current MIS capacity, their requirements, and what level of centralization of the information processes match their requirements. Each of the three information processes can be organized either more centralized or decentralized to meet the managers’ requirements. Ultimately, the MIS can be a totally centralized, decentralized or hybrid system.
7
1.4 Limitations
This thesis excludes several factors. To start with, this thesis leaves out the organizational point of view about the MIS. The focus is on the individual perceptions of employees towards the operations of the MIS. Therefore, several contingencies which often have been related to the MIS structure are omitted. Weill and Olson (1989), who summarized dozens of articles regarding MIS contingencies, found the following factors: the strategy, organizational structure, size, environment, technology, and nature of tasks. Since the target of this thesis is the manager within the sub unit of Air France KLM, only the nature of tasks is involved and elucidated. This contingency determines how the MIS is used and interpreted by the managers in different task circumstances (Mathiassen and Sorensen, 2006). The other contingencies do not influence the managers’ perception on the MIS nor the usage of the processes. The strategy, structure, size, environment, and technology do not change the way managers think about the functioning of the MIS. The managers’ requirements are based on the task circumstances; different situations call for diverse MIS processes (Lee and Leifer, 1996).
Another limitation is the exclusion of several variables which are often involved in the research about MIS performance. According to Weil and Olson (1989), there are four variables which can influence the MIS usage: the way of management or decision-making, implementation, MIS structure, and development or technical aspects concerning the MIS design. This thesis only focuses on the MIS structure and leaves out the rest of the variables. The way of managing or decision-making, for instance hierarchical or flexible, is not involved since this is already determined for the sub unit at Air France KLM. Decision-making at Air France KLM occurs centrally and will not change anytime soon. Therefore, this management variable is out of scope. The implementation of a new MIS is excluded as well. This thesis does not suggest a new system, but focuses on the execution of the current MIS and information processes. Consequently, there is not any focus on the development or technical aspects of the MIS either.
Futhermore, the cultural aspect is excluded as well. Within the sub unit of Air France KLM different cultures coexist. The perception about the usage and usefulness of the MIS could differ per culture. However, this is not important to discuss since the target of this thesis is the individual manager within the sub unit. More specifically, the perception of each manager is individually investigated. The investigation of the cultural background and its relation to the MIS usage is irrelevant to discuss since this would overlap the individual scope. More specifically, the individuals already provide more detailed information than the cultural background does.
8
1.5 Research Questions
To achieve the goal of this thesis, several questions have to be addressed. These questions are answered through both a literature review as well as a case study. Knowing the objective and limitations, the subsequent research question needs to be addressed to find the answer on the main question:
What level of centralization of the MIS processes matches with the managers’ requirements based on their task circumstances within the sub unit Air France KLM Asia Pacific Passenger’s Business?
Different requirements call for diverse MIS processes. It is difficult to establish one optimal MIS. This thesis tries to find a harmonized set-up to comply with the diversity of managers’ requirements. It regards what (de)centralized MIS features or capacities match with what managers’ requirements based on the nature of tasks. The different task circumstances apply for diverse information processes and characteristics. Success factors should indicate what MIS capacities are important and how these could successfully match with the managers’ requirements.
The main research question is eventually answered by investigating several subquestions. All questions are answered through scrutinizing related literature. The subquestions are:
1. How is the Management Information System structured? 2. What determines the MIS and its information processes? 3. What factors determine a successful MIS?
4. What are the capacities of centralized and decentralized MIS processes?
5. How should the MIS processes be structured to comply with the managers’ requirements? The case study addresses question one and five as for the sub unit of Air France KLM. Moreover, the answers of the other sub questions, based on the literature review, are applied to the sub unit to discover the state of the present determinants of the MIS, to check whether the managers are satisfied with the system and what should change, and what level of centralization is and should be maintained.
1.6 Research Methodology
9
supportive MIS processes for the managers. For instance, some managers are more concerned with computing information, while other focus on analyzing information.
The analysis of tasks is examined with a system and interaction approach to find the fit between managers’ requirements and MIS capacities. The system approach refers to a coherent concourse of different task circumstances, MIS designs, and characteristics. The interaction approach involves the managers’ requirements in relation to the task circumstances and alternative set-up of information process. These perspectives are chosen to show the relationship between dynamic task circumstances, managers’ needs, and required information processes.
The applied literature comprises articles from the business field, including literature on mainly management and information systems. Since the management information system is discussed in information technology (IT) journals as well, several articles are included from this part of the business field as well. The management and information journals elaborate on the MIS and its information processes, task circumstances of individuals, alternative information characteristics, MIS success factors, centralization and decentralization of the MIS, and the relationship between these subjects. Therefore, the management and information system literature is used to answer all sub questions. Addtionally, the few IT journals are involved to give extra information on the MIS structure which is necessary to examine the information processes. This part of the business research is used to answer the questions about the MIS structure, determinants, capacity, and mutual relationships. This refers to all the sub questions except the third one. It is important to see how an MIS is built up and works to realize how the information processes and its output should be organized. This is essential to find a successful match between the MIS requirements and capacities based on theory. This thesis connects the mentioned factors to determine the requirements. The related factors conclude what capacity is needed to satisfy the MIS users. Eventually, the literature review should give fruitful insights in how to organize the production, usage, and sharing of management information based on the managers and its task circumstances.
The theorectical contribution is used to develop a conceptual framework to find the successful match for the sub unit. The theory is applied in the case study to see whether the current management information processes are in line with the literature review. The case study shows how the information processes are currently applied. More specifically, how is the management information produced, used, and shared by the managers? Morevover, the case study discovers how the management information could be successfully processed regarding the managers task circumstances. What the information processes should be capable of determines the level of centralization of the MIS (Lee and Leifer, 1996).
10
unit. According to Blumberg et al (2008), the case study is a well-recognized and often applied methodology amid management studies to answer research questions. Besides, it is applicable to exploratory case studies such as this one. This case study is explorative since it investigates the success characteristics through a set of subjective data. The data collection combines information from observations, archives, and surveys (Eisenhardt, 1989). According to Yin (1984), a case study is an empirical inquiry which relies on a triangular data set. According to Guion et al (2002), triangulation is a method for qualitative research to create reliability and validity through analyzing the research question from multiple perspectives. The data collection is used to discover what level of centralization of the information processes is currently used and what should be used according to theory and the managers.
Initially, the observations show real life experiences of the MIS processes. The observations are made during the internship at the headquarters of KLM in Amstelveen from October 2011 till April 2012. Throughout these months, the activities of the managers at the Air France KLM Passenger’s Business unit are observed. This regards the day-to-day activities, meetings, trainings, and weekly conference calls with MIS specialists, regional, and area managers.
Morever, the documentation gives a fruitful insight in the applied information. The documentation includes management information overviews or reports and presentations of meetings and trainings. Furthermore, the survey is applied to verify the managers’ needs and opinions regarding the MIS processes. A survey is preferred above interviews since the managers are widely dispersed over Amsterdam, Paris, and the whole of Asia. Therefore, they are difficult to reach due to distance and time differences. Besides, the needed data is easier to collect and quantify via surveys including scales and straightforward questions. There is no need to interrogate the managers on certain matter since the questions are clear. Therefore, a survey sustains. The data collection entails both qualitative and quantitative material. It focuses on what the managers consider to be important. The task circumstances and success factors are questioned qualitatively but scaled to quantify the managers’ opinions and enable to compare the answers. The implemented scale is the Likert-scale which relates qualitative measures such as ‘strong’ and ‘weak’ to a grade from one to seven. The expectation is to find aligned results from the different sources since it is a coherent data collection focusing on related aspects.
Eventually, the data collection is used to answer question five. However, to find the match, all of the questions and answers based on the literature review are examined in the case. The case study shows the MIS structure, determinants, success, and level of centralization of the MIS. Hence, the achieved match for the sub unit is partially based on theoretical findings and partially on the data collection.
1.7 Relevance
11
processing and sharing of information. From my experience, new centralized or decentralized MIS are often introduced and implemented without consulting the managers and taking their requirements, opinions and task cirumstances into consideration. Consequently, the MIS often does not fully cohere with the managers’ requirements and are not supportive to their activities. This thesis shows what level of centralization of the MIS is required in certain situations. It shows what system and information requirements and capacities are needed to meet the managers’ perception on the content and usefulness of the MIS. It combines different frameworks of a broad range of literature. Where other literature primarily focuses on either the MIS capacities or requirements, this thesis combines both aspects since they are inextricably linked. Previous literature has not involved the managers and their task circumstances in relation to the level of centralization of the MIS. Moreover, the success factors have not been linked yet to the centralization level of the MIS either. Besides, converging subjects have not been discussed often, especially not recently, while it is still a contemporary issue. The literature review and the subsequent conceptual model are useful for organizations and sub units to discover how to organize their MIS processes, either centralized, decentralized, or in a hybrid way, based on the managers’ task circumstances. It is useful to show what MIS processes managers may require and should have according to theory. This set-up and combination of approaches, level of analysis, and involved theory to examine a MIS has not been used before. Since it is still a great deal to find the right level of centralization of the MIS while accounting for the managers’ needs, this thesis gives fruitful insight into essential aspects which need to be involved.
1.8 Research Design
This thesis uses a literature review and subsequently a case study to find out what level of centralization of the MIS processes matches with the managers’ requirements based on their task circumstances. Chapter 2 contains the literature review which is used to address all sub questions. The literature review contains the elucidation on numerous aspects of the management problem to understand and find the match. Chapter 2.2 starts with defining the main aspects of the research question: the MIS, information processes, and centralization of the MIS. This is essential to indicate how these concepts are interpreted and treated alongside this research.
Additionally, the sub question ‘How is the Management Information System structured?’ is answered in section 2.3. It is important to scrutinize the MIS and its processes to be able to address what parts could be centralized or decentralized based on the task circumstances. It construes the MIS processes and underlying activities which is useful to discover how the information is processed and should be processed.
12
and related variables are shown. Subsequently, the information characteristics and its relationship with the contingency variables are examined. The MIS success factors and its connection to both the variables and characteristics are elucidated. The success factors regard the managers’ perception on the contentment and usefulness of the MIS. This information is necessary to address the following sub question: ‘What factors determine a successful MIS?’ Moreover, this section shows how the tasks circumstances, information characteristics, and success factors are related to describe different requirements.
Moreover, in section 2.5, the other side of the potential match is discussed: the MIS capacities. The advantages and disadvantages of both the centralized and decentralized MIS processes are given. The features of centralized and decentralized MIS processes are shown. This section elaborates on the question: ‘What are the capacities of centralized and decentralized MIS processes? The results of this section are needed to be able to compare the managers’ requirements with the capacities of centralized or decentralized information processes.
In section 2.6, the managers’ requirements and MIS capacities are compared and matched. It addresses the question: ‘How should the MIS processes be structured to comply with the managers’ requirements?’ This section needs to show what level of centralization the MIS processes should entail to meet the managers’ requirements. Moreover, the level of centralization is related to the task circumstances, information characteristis, and success factors. This is necessary to conclude how the MIS processes should be established in certain circumstances.
In section 2.7, the conceptual model gives an overview of the factors of the previous sections. It shows the relationships among the factors and how these connections lead to a match between the managers’ requirements and centralized or decentralized capacities.
Chapter 3 entails the case study based on the conceptual model and data collection. Section 3.2 shows the results. More specifically, the data collection is related to the MIS requirements and capacities to find the match according to the managers of the sub unit of Air France KLM and its task circumstances. Furthermore, in section 3.3, the results are analyzed. The practical findings are compared with the theory of the literature review. This section shows whether the theory is in line with the currently applied MIS processes and if the managers are satisfied with the MIS. The task circumstances, information characteristics, and success factors are examined via the data collection to determine the managers’ requirements and the level of centralization of the MIS processes. This should result in successful centralized, decentralized, or hybrid information processes. In section 3.4 is discussed what should possibly be adapted to the MIS processes to find an appropriate match and the right level of centralization.
13
Chapter 2 - Literature Review
2.1 Introduction
The subsequent literature review contains articles from business journals regarding management, information systems, and information technology. This range of business articles is necessary to address the management problem. The literature from management and information system journals pertain to all sub questions. This literature relates to the task circumstances, information attributes, success factors, MIS structure and centralization, and mutual relationships of these subjects. The IT articles provide further information on the MIS structure, its determinants and capacity, level of centralization, and relationship with the managers’ requirements. Moreover, the articles cover a broad time span from 1973 till 2011 since there are not many recent articles on related subjects. The scarcity of literature on this matter is dispersed over time.
The applied literature is divided over six sections. Respectively, the terminology, MIS structure, requirements, capacities, match, and conceptual model are discussed.
2.2 Terminology
This literature review is initiated by defining the most common concepts of this thesis. This is essential to avoid possible confusion through the diversity of definitions around single concepts. Moreover, it is necessary to frame the concepts to direct examinations and discussions to answer the research questions. Therefore, the set terms in this section are maintained throughout this thesis to avoid ambiguity.
2.2.1 Management Information Systems
14
information about the organization’s activities to regulate the activities. This leads to different amounts, formats, and sorts of information per level. The MIS and its information processes should be able to comply with diverging requirements. Hence, a MIS is successful when there is a match between its capacities and the requirements of the users. The requirements imply what kind of information users need, but also in what way the information should be delivered. This deliverance of information entails the characteristics and quality of the MIS. As mentioned before, this thesis focuses on the information requirements by looking at the deliverance of information rather than investigating the preferred figures and formulas. The supportiveness of the system is questioned. According to Weill and Olson (1989), an appropriately designed MIS provides the required support to the organization’s information processes. The type of support depends on the the nature of tasks of managers. A well-structured MIS needs to benefit the performance of task executions. Since it is impossible to measure MIS performance quantitatively, the success of a MIS is determined by the perceived usefulness and user’s satisfaction which is described more thoroughly afterwards. Both usefulness and user satisfaction are mainly determined by the quality of information, the system, and its usage (DeLone and McLean, 2003). Therefore, these factors determine the information process’ requirements. Next to the requirements, the capacity of the system is elucidated. The capacity of a MIS differs per system. According to Mathiassen and Sorensen (2008), information systems can be either adaptive, collabarative, networking, or computational minded. Each interpretation matches a certain nature of tasks. Moreover, each system entails different capacities or features to deal with the nature of tasks and support managers. The differences are based on standards or flexibility which imply centralized or decentralized systems. All of these MIS styles interpret the production, usage, and sharing of information differently. The idea is to match these capacities to the requirements to satisfy managers amid different circumstances.
2.2.2 Information Processes
15
preparing of data to make it useful to the MIS users. Hence, there is data, information, and relevant information. Mason and Mitroff (1973) define relevant information as knowledge intented to take successful actions. This regards the second stage of information processing: the usage of information. Whether an action or task is executed successfully is disputable. This thesis focuses on the contribution of successful information processes to enhance task execution rather than to measure the actual benefits of the accomplished tasks. The success of the managers’ operations is expressesed as user satisfaction which is discussed further on in section 2.4.6. The results of task execution, such as profitability, are not of interest for this research. The execution of tasks depends on the performance of the information processes. Chang et al (2003) supports these functions of the MIS, however, he involves another function of the information process. To start with, the MIS should summarize the sub units’ operational performances. This regards the production of information about the business. Secondly, the MIS should allow for recording, processing, analyzing, and displaying information to execute tasks. This refers to the usage of information. The extra or third function of information processing is the provision of information to accomplish tasks. This is in line with Lee and Leifer (1992), who emphasize the production, usage, and sharing parts of the MIS. The implementation and execution of the three information processes do all depend on the information capacities. The three aspects should be based on the requirements of MIS users to satisfy the managers. Hence, the information capacity should be able to facilitate these requirements.
2.2.3 MIS Centralization
The production, usage, and sharing of information differ per MIS and level of centralization. According to Beck (2010), MIS centralization implies that one specific unit or MI team allocates identical information its employees. Van Zandt (1997), supports this by stating that centralized information processing refer to one MIS unit which processes and distributes equal information. For instance, centralized information processes imply one database including all the facts and figures and a central sharing system which provides the same information to each manager. The production, usage, and sharing of information is rigid and arranged by a central MIS unit or specialists. The MIS unit sets standards which restrains the managers’ ways of operating. There is much control over the information processes. This could be useful with specific task circumstances. Other conditions could however demand more flexible or decentralized systems. According to Beck (2010), MIS decentralization gives individuals autonomy over their data and information exclusive of major restrictions or standards. The information processes are less confined and there is less control over the resources. Managers have more options to deal with fluctuating tasks. The information process activities are more adaptive to changes and desires.
16
centrally. This means that all of the data is stored into one datawarehouse or database and is equally provided to all of the users. The information is uploaded into the users’ databases, they do not have to produce any information, and can start focusing on the usage of information immediately. The production of information is centrally executed. Once the information is produced, the MIS users can apply the information. The users rely on the same data, information, tools, and interpretations which are set by the central unit.
On the contrary, a decentralized storaging system entails numerous information providers. The sub unit or organization applies different databases. The provided data is customized to the MIS users. The information is not equally produced for each user. Consequently, the decentralized systems and its users work with different information and processes (Mozkowitz and Murnighan, 1975). Besides, the users are involved in both the production as well as the usage cycle due to the decentralization of information processes. Since the data is not similar among the MIS users, they have to produce information overviews themselves to support each other. The MIS users have different options and tools to process information. Moreover, there are not rigid standards to align interpretations.
Moreover, the sharing of information can also be centralized or decentralized. For instance, the reporting and exchanging of information can be executed centrally. A sub unit could implement a centralized sharing system to send out standard reports to everyone on a regular basis. Each MIS user receives the same information. In contrast, the individuals could decide to exchange customized information overviews and reports themselves. In this case, the MIS users decide what and when to share. Besides, the MIS users can choose to whom they send information and from whom they would like to receive reports.
Furthermore, the degree of centralization of MIS control is essential when structuring the information processes. This regards the number of hierarchical levels and authorized users. Sub units could decide to have one or more authorized users who can control the MIS activities including the database processes and sharing of information.
2.3 MIS Structure 2.3.1 Introduction
The subsequent paragraphs elaborate on the question: How is the Management Information System structured? The framework of the MIS is based on the model by Barros (1981). This framework gives an accurate view of the set-up and related tasks regarding the production, usage, and sharing of information. Moreover, this view is useful to gradually investigate the MIS alongside relevant literature and the case study.
2.3.2 Composition MIS and Information Processes
17
necessary to explore the information processes to create more references to investigate and relate to the task circumstances, information characteristics, and success factors. The overview is based on a match between the models and interpretations of Barros (1981) and Mathiassen and Sorensen (2008). The MIS and information processing activities pertain to three managerial levels of a sub unit or organization. More specifically, the operational, tactical, and strategic positions are included. The different positioned managers exchange information to execute tasks, to manage diverse activities, to comprehend the environment, and to check performances alongside the MIS. The processing of information per level can be different due to authority, accessiblity, and needs. It is possible to interpret the MIS in different ways. Figure 1 shows the traditional usage of the MIS, including the three managerial levels and the related MIS processes.
Figure 1: Traditional information processing model
18
Finally, the computed information is provided to the other managers. The provided data by the operational managers is returned, however, transformed and computed into useful information. Besides, the strategic managers can request this kind of information as well. However, since this information is not modelled nor analyzed it is not always helpful to strategic managers. Top level managers prefer summarized analysis including straightforward conclusions for decision-making. Basic information is not sufficient in this case since it still needs to be processed before it is applicable. Therefore, next to the sharing of this basic information the usage cycle of information is involved. This part is more difficult and time-consuming since conclusions need to be drawn from the basic informative overviews. The information needs to be sorted to make it applicable for task execution. According to the traditional usage of the MIS, tactical managers are responsible for these activities. They make structured summaries of the performances which are applicable for the execution of tasks. Besides, the tactical managers can exchange these files to discuss certain issues and draw aligned conclusions. The first step in this part of the information process is the accumulation of data. Data needs to be aggregated into overviews. The level of aggregation depends on the individuals’ requirements and nature of tasks. This is discussed in section 2.4.4. According to Barros (1981), the accumulation of information incorporates the data and information into models and overviews which are necessary to explore developments in the business performance. This exploration or analysis of data is the next phase of this process. The outcomes of the models and the overviews have to be analyzed to elicit certain trends. This is important to anticipate and improve performances. Furthermore, the analyzed figures or qualitative material initiates the calculating process. The material is useful to make calculations or estimates about developments and prospects. Altogether, these steps lead to the final phase: informing the strategic managers. The processed information is shared among the strategic and tactical managers and occasionally with the operational level. The top level managers receive processed information to make decisions and give feedback to the tactical managers.
19
2.4 MIS Requirements 2.4.1 Introduction
This section elaborates on the sub question: What determines the MIS and its information processes? The previous section showed the existence of numerous MIS capacities. This also applies to the set of requirements. Managers’ requirements diverge alongside different task circumstances. This section discusses this specific contingency of the information processes. It shows different contingency variables which determine the task circumstances and, consequently, the information requirements. These requirements entail the information characteristics and MIS success factors. Initially, these concepts are explained. Subsequently, the relationship with the contingency variables are discussed. Altogether, this set-up of factors is chosen since there are several significant relationships between them (Mangaliso, 1995; Lal and Hassel, 1998; Choe, 1998; Chang et al, 2003). However, none of these authors combined nor related all factors which are discussed in this thesis. The integration of the different connections is necessary to develop the conceptual model and to investigate the MIS match both in theory as well as in reality.
2.4.2 Contingency MIS
20
explore what related contingency variables exist. When investigating the previous literature regarding these variables, this leads to two commonly applied and related factors: task uncertainty and equivocality. Both variables determine the nature of tasks regarding the information processes (Ginzberg, 1980). Besides, both factors are related to information characteristics and MIS success factors as well. These relationships are discussed during this section as well.
2.4.3 Contingency Variables
21
which is less error-prone and manipulatable than the other cycles. Apart from that, data or information could be inserted wrongly into the systems or mismatch between the users’ systems. Within the other cycles information is more devoted to the managers. They compute and analyze information based on subjective perceptions. Managers could make errors during analyses or manipulate information. Hence, the usage and sharing cycle are more vulnerable to errors and misuse which could damage the data integrity. Besides, the MIS users face uncertainty when it is hard or impossible to find information to make accurate predictions (Milliken, 1987). This shows the usage cycle is highly dependable on the production cycle, user friendliness of the MIS, and the skills of the managers. Primarily, when managers face uncertainty, they need more information to be produced. This new information should be easy to process to be able to draw conclusions and execute tasks. The level of new information and how the processes are organized fluctuate with the amount of uncertainty. Therefore, the MIS should be structured in that way it can deal with the amount of uncertainty. MIS users diminish uncertainty through obtaining and analyzing more information. They benefit from more flexible information processes to be able to produce new information (Mathiassen and Sorensen, 2008). Managers who face low uncertainty can rely on basic information processes. When there is high uncertainty, the MIS could use more regular reporting, guidelines, and communication such as conference calls and meetings (Daft et al, 1987). Higher uncertainty asks for extra information which means an increasing demand of, for instance, customized reports and information packages. Hence, the balance of the production and usage cycle shifts with the level of uncertainty. A MIS should support the production and sharing of information when managers face high uncertainty. During low uncertainty, the managers are better off when the MIS simplifies the usage of information (Mathiassen and Sorensen, 2008).
22
Next to uncertainty, the managers face a certain level of task equivocality. Equivocality or ambiguity is the presence of various interpretations about information and processes which instigates misapprehension, disagreement or lack of understanding (Daft et al, 1987). These conflicts disorder the information processes; they cannot be executed in a straightforward way. Besides, it is harder to obtain the required information to fulfil a task. Managers encounter time-consuming efforts to solve the ambiguity. Since information and processes are not aligned, managers could have different interpretations. Therefore, more standards need to be introduced to diminish the ambiguity. The exchanged information needs to get aligned to reduce conflicts due to different data and defintions. Managers should increase interaction to get the interpretations aligned (Mathiassen and Sorensen, 2008). The equivocality should decrease when the interpretations match. Furthermore, according to Daft and Macintosh (1981), the equivocality is influenced by task analyzability which concerns the structure of information. The structure refers to the applied formulas, definitions, and formats. Understandable and aligned structures simplify the sharing of information since the analyzability is higher. The analyzability also refers to the difficulty of the information process during the execution of tasks (Chang et al, 2003). According to Choe (1998), task difficulty is the ability to evaluate information and tasks. More difficult tasks lead to uncertainty since it is harder to get the right information and draw conclusions from it.
Altogether, these concepts determine whether the tasks are routined or not. The higher the uncertainty, equivocality, variability, and non-analyzability are, the less routined are the tasks and information processes (Daft and Lengel, 1986). This situation demands extra effort of the MIS users and suitable information process capacities. The right MIS can alleviate the circumstances. Information processes can reduce both the uncertainty and equivocality if they suit the managers’ needs and activities (Lewis, 2004). The previous paragraphs showed that uncertainty is related to the production and usage cycle. It is about the interpretation and organization of the database and related programs. It regards the computerized systems including the data and information such as revenues and costs. The level of uncertainty determines how the database should be organized. The level of centralization of the databases depends on the availability and reliability of information next to the variability and difficulty of tasks. This implies that the centralization of the MIS is based on the uncertainty and equivocality. For instamce, when the uncertainty rises, the systems should comply with the changes by being more flexible. The sharing of information is also determined by the level of the contingencies. For instance, higher ambiguity asks for more standards to avoid conflicts.
23
have to indicate the nature of tasks and the individual perception upon them. Different combinations could be implemented. Since all factors are related it should be possible to find matches for different circumstances.
2.4.4 Information Characteristics
According to Chang et al (2003), there are three functions of the MIS to organize information. Primarily, the MIS should summarize data of the sub units’ operational performances. Secondly, the MIS should facilitate the recording, processing, analyzing, and displaying of information to be able to execute tasks. Ultimately, the MIS needs to provide the required information to its users to accomplish tasks. These three functions can be addressed in different ways. The previous section showed that the managers’ requirements differ alongside the task circumstances. Managers apply for different sets of information with diverging characteristics. This thesis focuses on the information characteristics given by Chenhall and Morris (1986) since these attributes have often been used to describe the information process. Besides, these characteristics are correlated to the contingency variables, MIS success factors, and information capacities of the MIS. These relationships are eventually fruitful to set up a conceptual model for finding the right design. Before describing these relationships, the information characteristics are elaborated. The aimed information characteristics are: scope, timeliness, aggregation, and integration.
The first characteristic, the scope of information, concerns the focus, way of measuring, and time-horizon of information processing. The focus can be centred on either internal, external or both kinds of information. Internal information refers to facts and figures about the operational activities of an organization itself. External information includes data from outside an organization such as market movements. The way of measuring could be executed quantitatively or qualitatively. A quantitative way of measuring focuses on figures while a qualitative approach centres on written facts. The chosen time-horizon of information shows from what period the data is picked. Managers can use past-, present-, and future-oriented information to execute tasks. Altogether, the scope of information can be either narrow or broad. A narrow scope of information refers to internal, quantitative, and historical data. On the contrary, a broad scope of information concerns external, qualitative, and future-oriented material.
24
The third factor is aggregation of information. This regards the choice whether to accumulate information over a period of time or different markets. Sometimes MIS users need all-embracing overviews while others sustain with less aggregated information to avoid time-consuming allocation. The final factor is integration of information. This concerns whether or not the information overviews of different functional units should be integrated. This could be interesting when different functions within one organization have mutual concerns. Altogether, the information characteristics can be used in different combinations to suit the circumstances of the tasks (Bouwens and Abernethy, 2000). However, the design of the MIS should be compatible. Besides, the combination of characteristics might have to change due to altering circumstances and positions within the sub unit. According to Gorry and Morton (1989), the subsequent changes of the information requirements occur if the managerial level changes:
Information characteristics Operational level Management level Strategic level
Scope Internal External
Scope Narrow Broad
Aggregation Low High
Timeliness Historical Future
Timeliness High frequency of use Low frequency of use
Table 1: Required information characteristics per managerial level
The requirements per level are straightforward. At the strategic level managers require broader and more aggregated information due to a wider span of control. Besides, their focus is on futuritive rather than on historical figures. Moreover, the strategic managers do not use the processed data for a long time since they need newly produced information constantly to be able to accomplish its tasks. Lower level managers need less accumulated data, but maintain a higher usage rate of information since they have clearer and less varying tasks. The case study examines whether these theories are in line with the Air France KLM sub unit and its managers.
2.4.5 Contingency Variables and Information Characteristics
25
under higher uncertainty the production cycle increases relatively to the usage cylce. MIS users such as managers need more information than already available. This causes extra production of information. In this case, MIS users should be careful to avoid an overload of information. In contrast, under low uncertainty the MIS users can rely on the available information. However, abundance of information could distract and disorder the execution of tasks. Managers could lose track in the excess of information. In this situation, the focus should be on the usage rather than on the production of information. Therefore, the nature of tasks should be examined constantly to mantain and retain the right balance between the production and usage of information. This is essential to avoid either a shortage or excess of information.
Refering to equivocality, Choe (1998) found significant and positive relationships between high task non-analyzability and broad scope, aggregated, and timelier information. Whereas timelier means futuritve information which is used on a lower frequency. The produced information is only shortly applicable since the variance of tasks require regularly new sources. The created overviews cannot used more frequently since these are quickly outdated. The MIS users only benefit from these attributes when they are highly involved in the implementation and adaptations of the system. The system needs to be open to changes after the installation to comply with potential alterations of the circumstances. On the contrary, rigid systems are more fortunate and applicable in providing narrow scope and disaggregated information due to less features. Again, the user participation is important to create synergies between the MIS and its user.
Daft and MacIntosh (1981) state that under high task equivocality the problem and its consequence are not clear. Hence, the MIS users require more aggregated information to get a better overview of the problems and the related effects. Chang et al (2003) supports this, however he refers to task analyzability instead of equivocality. When the non-analyzability of tasks increases, managers need more aggregated information including a wider range of time periods and markets to accomplish diverse tasks. According to McDonough III and Leifer (1983), the level of routinization regarding tasks is shown by the occurrence of exceptional events and the degree of analyzability. Routine tasks have little exceptions or uncertainty and high analyzability due to low equivocality. On the contrary, non-routine tasks are characterized by irregularity and diverging interpretations which lowers the analyzability and increases equivocality. Altogether, the information sharing requirements increase alongside the rising uncertainty, variability, equivocality, and non-analyzability (Lee and Leifer, 1992).
2.4.6 MIS Success Factors
26
27
Due to probable disparity among the priorities of MIS users, another model or set of factors is involved as well. This extends the examination possibilites of the MIS. A larger range of success factors will provide a more accurate view of the perception of the MIS users on the usefulness and satisfaction of the system. This is necesarry to assess the success of the MIS. According to Ginzberg (1980), the success of a system within a sub unit depends on the fit of the individual needs and system characteristics. To find this match and combine the factors, the success model by DeLone and McLean (1992) is applied. This model includes the system, information, and service quality, next to the importance, usage, and usefulness of the MIS. These factors are interrelated to eventually determine the user satisfaction of the system. Since all of these factors are enlaced in the this success model, this provides a good combination. Besides, the discussed information characteristics are related with and intertwined in the factors of the model as well. This is elucidated later on. The basic success model is shown down below:
Figure 2: MIS success model
This model focuses on the impact of the MIS on the individual MIS users which will eventually influence the whole sub unit or organization. The first four blocks are used to assess and explore the individual opinions on the MIS which indicates its impact on the execution of tasks. The system and information quality determine the usage and user satisfaction of the MIS. Altogether, the first four blocks of the model elaborate on the MIS and its impact on the individual MIS user. This impact influences the overall effect of the MIS on the sub unit or organization. It is an appropriate model to check the opinions of MIS users on the applied information processes. The model has often been used in research which shows confidence in these factors. Besides, Seddon and Kiew (1996) tested the model and found significant relationships between the factors and with the individual impact. Scott (1995), also tested the success model of DeLone and McLean (1992). He found positive and significant relationships between system and information quality next to the MIS usage, user satisfaction, and the individual impact on the execution of tasks. Furthermore, the factors are related to the mentioned contingency variables and information characteristics. These connections are discussed in the section hereafter. These relationships are based on several articles in which these connections are tested. Moreover, the DeLone and McLean model from 1992 is extended by several dimensions to enlarge the analyzability of the success of the MIS. In 2003, DeLone and McLean augmented the model by including the factor service quality. Another factor is included on behalf of Sabherwal and
System Quality
Individual Impact
28
Chowa (2006) which regards the perceived usefulness of the MIS. Finally, the importance of the system is incorporated. These three factors are also positively and significantly correlated to the other factors. Before discussing the connections, the success factors of the model are explained:
- System quality: implies the value of the MIS and information processes itself. The measurement of system quality entails many factors such as data and system accuracy, data currency, database contents, ease of use, ease of learning, convenience of access, usefulness of the system, system flexibility, system reliability, system sophistication, integration of systems, systems efficiency, and response time (DeLone and McLean, 1992)
- Information quality: regards the value of the output of the MIS processes. According to DeLone and McLean (1992), the following factors are of interest: importance, relevance, usefulness, informativeness, usableness, understandability, clarity, format, content, accuracy, precision, conciseness, suffiency, completeness, reliability, currency, timeliness, and comparability
- Service quality: denotes the support of specialists and designers to usage and maintenance of the MIS. When MIS users face certain problems during the information processes, they should be able to get assistance to solve the issues. The service quality can be measured on behalf of the following factors: reliability, responsiveness, assurance, and empathy of MIS consultants (Gorla et al, 2010)
- System importance: entails the pregnancy of the MIS to its users. Seddon and Kiew (1996) added this factor to the success model of DeLone and McLean (1992) to get a more accurate insight in the opinions of MIS users on the usefulness and user satisfaction of the MIS
- Usage: concerns the required effort to use the MIS successfully. It includes factors such as the needed amount of time to process information and the regularity of usage (Seddon and Kiew, 1997)
- Perceived usefulness: expresses whether the MIS users benefit from the system or not. It regards the level of productivity which is either enhanced or obstructed by the MIS. It also refers to the performance and effectiveness of the execution of tasks (Seddon and Kiew, 1996)
- User satisfaction: states whether the MIS corresponds with the requirements of its users (Sabherwal and Chowa, 2006)
Altogether, the interpretation of these factors lead to a certain impact on the individuals. They affect the execution of tasks. The next paragraph shows the relationships between the success factors themselves to discover the influence of the MIS on the individual users.
2.4.7 Contingency Variables and MIS Success Factors
29
availabilty of information but also desire assistive information processes. This combination could lead to successful operating conditions. Besides, the information, system quality, and the perceived importance of the system are positively and significantly related to the usefulness of the MIS. More specifically, this means the managers consider a well-designed and useful system to be important. Ultimately, these factors have a great share in creating user’s satisfaction of the MIS. Apparently, the MIS users attach great value to these qualities. Therefore, it is interesting to consider what factors influence the information and system quality the most. According to Nelson et al (2005), information quality is strongest related to accuracy, followed by respectively completeness, format, and currency of information. These factors determine the information quality and, therefore, have to be adjusted to the managers’ necessities. The system quality is mainly influenced by respectively the reliability, accessibility, flexibility, integration, and response time of the MIS. These factors should comply with the nature of tasks and corresponding managers’ needs. Moreover, the relationships of both these information characteristics and success factors are positively and significantly related to user’s satisfaction. Therefore, when these factors meet the needs of managers, the MIS is more appreciated and beneficial. Gatian (1994) adds that harmonizing capacity and requirements stimulate the efficiency of information processes and the exectution of task by MIS users.
30
quality which means broad scope, timely (futuritive data, high frequency of production, and low rate of usage of information), and aggregated information. However, when the users face low uncertainty, the information should have a narrow scope, be less aggregated, and be less timely (historical data, low frequency of production, and high rate of usage of information) to be sufficient and satisfy the managers. Additionally, Chang et al (2003) found that when the MIS users face high task variablity, they are better off with broad scope information to increase the MIS performance. On the other hand, when there is low uncertainty, the MIS user benefits from narrow scope information to increase MIS performance.
Due to the found and incorporated relationships it is possible to extend the model from figure 2. This extended model is shown in figure 3. This representation is used during the rest of the literature review to explain relationships with the other information requirements and capacties. Finally, this model is incorporated into the conceptual model for the case study. The success factors are used to gain feedback on the MIS of the sub unit of Air France KLM. This regards the individual perception of the MIS users on the system. The perceptions give an overview of what part of the MIS is or is not operating successfully. Each block within the model of figure 3 is elucidated. Eventually, the underlying variables of the system, information, and service quality can be adjusted to increase the perceived usefulness and user’s satisfaction to create a more successful MIS impact on the individuals.
Figure 3: Extended MIS success model (DeLone and McLean, 1992; Seddon and Kiew, 2007)
2.5 MIS Capacities 2.5.1 Introduction
31
requirements to satisfy the users. Harmony can be established through three different systems: the centralized, decentralized, and hybrid system. The hybrid MIS is a mixture of centralized and decentralized features. The two main features are the database and the sharing system. The database contains data or information to analyze the performances. The sharing system organizes the exchange of information among the MIS users to discuss and address certain issues. It is important to install and maintain the right MIS features to guarantee the required information capacity. Too little capacity causes information deficiency which complicates or even obstructs the execution of tasks. On the contrary, too much capacity causes information abundance which distorts the completion of tasks due to lack of focus (Chang et al, 2003). The right set-up obliges the most appropriate level of centralization of the MIS. The organization or sub unit should consider whether to centralize or decentralize the information processes (Mozkowitz and Murnighan, 1975). This thesis assays both the features of a centralized and a decentralized system to simplify this consideration. The next section matches these features with the requirements based on the level of uncertainty and equivocality. Previous literature found relationships between the nature of tasks and the MIS design. For instance, Ginzberg (1980) confirms the influence of the condition of tasks on the MIS design. Furthermore, Chenhall and Morris (1986) show the effect of the tasks’ circumstances on the MIS structure and the information characteristics.
Complementary, a case study is used to show what information characteristics and success factors are important to the MIS users in specific situations. The level of uncertainty and equivocality during the execution of tasks is investigated to describe these specific situations. Altogether, these three aspects are matched with the most suitable system: centralized, decentralized or hybrid. The next two paragraphs elaborate on these decisions by looking at the (de)centralized capacities of a MIS. The section thereafter matches the requirements with the capacities to resolve when a centralized, decentralized or hybrid system needs to be implemented.
2.5.2 Centralized MIS capacities
