Allocating Ownership to Manufacturing
Information: A Case Study in the
Aerospace and Defence Industry
A. Mulder
Master Thesis
University of Groningen: Faculty of Economics and Business MSc Technology and Operations Management
2 Master thesis
Allocating Ownership to Manufacturing Information: A Case Study in the Aerospace and Defence Industry A. Mulder
Peizerweg 48, 9726JK Groningen A.Mulder.18@student.rug.nl A.Mulder1@newcastle.ac.uk Tel: +31(0)6 13741730
Student number RUG: 2200104 Student number NUBS: 140623242
December 7th, 2015
University of Groningen, Faculty of Economics and Business Master of Technology and Operations Management
Supervisor: prof. dr. ir. J. C. Wortmann Nettelbosje 2, 9747 AE Groningen Tel: +31(0)50 3633741
Newcastle University Business School
Master of Operations and Supply Chain Management Supervisor: dr. Y. Yang
3
Abstract
This study examines why companies are struggling with the allocation of ownership to information. While high quality information is essential for manufacturing companies in order to produce efficiently, few companies have information governance and ownership policies in place. Literature on information governance is still scarce, yet it has showed promising results in both theory and practice. Academia has however not focused on the accountability aspect of information governance, which will increase information quality by holding people accountable for the immediate correctness and consistency of information. Companies that have showed efforts to implement information ownership frequently encounter difficulties in practice. This research tries to identify why companies are struggling with the allocation of information ownership by performing a single case study in the Aerospace & Defence industry. Due to the highly regulated nature and the reliance on high quality information, companies in this industry are deemed to be interesting as a case. Three sub-questions have been answered which gave insight in the roles that are accountable for some level of information quality, the information flow, the assignment of ownership and the way accountability is handled within the case company. This study found that information ownership was allocated to specific people, rather than the roles these people fulfil. The findings show that ownership is automatically formed and assigned to the creator of information. The lack of rules and guidelines regarding information ownership resulted into ambiguity regarding the accountability when people other than the owner modified information. Therefore, further research is needed into the possibility of incorporating information ownership as a role in the governance policy.
Keywords: Information ownership, Information governance, Manufacturing
4
Management Summary
This dissertation focuses on the question why companies are struggling with the implementation of information ownership. Fokker Aerostructures produces myriad complex products for which a large range of high quality information is needed. Managing this information is of the essence in order to remain this quality. Currently several information management practices are in place, such as regulations about the proper use, creation and storage of information. Furthermore, the quality department plays a role in ensuring that the information is correct. Literature advocates the use of information governance practices and information ownership to assist in the management of this information. Information governance contains standards, guidelines and controls to guarantee the quality of information within a firm. The governance framework specifies roles to people that are accountable for information. These roles manage information access, standards, rules, formats and many more elements. By the assignment of ownership, as part of information governance, people can be held accountable for the immediate correctness and consistency of information.
5
Table of Contents
PREFACE ... 6 1. INTRODUCTION ... 7 2. THEORETICAL BACKGROUND ...10 2.1 INFORMATION QUALITY ... 10 2.2 INFORMATION GOVERNANCE ... 13 2.3 INFORMATION OWNERSHIP ... 17 2.4 ACCOUNTABILITY ... 21 2.5 RESEARCH SETTING ... 23 3. METHODOLOGY ...25 3.1 LITERATURE STUDY ... 263.2 CASE STUDY PROTOCOL ... 27
Overview of the case study project ... 28
Field procedures ... 28
Case study questions ... 29
Case study report guide ... 30
3.3 DATA COLLECTION ... 30
3.4 ANALYSIS OF THE FINDINGS ... 31
4 FINDINGS ...33
4.1 QUALITY ROLES AND INFORMATION FLOW ... 34
4.2 ASSIGNMENT OF INFORMATION OWNERSHIP ... 38
4.3 GUARANTEEING THE CORRECTNESS AND CONSISTENCY OF INFORMATION ... 41
5 DISCUSSION ...44
6 CONCLUSION ...51
REFERENCES ...52
6
Preface
This dissertation is written as the final assignment for the Master of Science degree in Technology and Operations Management at the University of Groningen and the Master of Science degree in Operations and Supply Chain Management at the Newcastle University Business School. By focussing on the common theme within these studies, a single thesis is written in order to fulfil the requirements for both. As both studies put an emphasis on operations management, I chose a subject that is closely related to this. Information is essential for successful operations management and the ownership thereof is a surprisingly difficult concept to grasp. Therefore it had immediately piqued my interest.
I enjoyed writing this thesis and learned a lot in the process. Spending six months on site at a company has given me great insight into how companies operate and helped me putting the theoretical knowledge gained through my studies into context. Therefore I would like to thank the people at Fokker Aerostructures who contributed to the establishment of this dissertation. I especially would like to thank Steven Hengeveld, Johan Nolle, Remko Schra and the team members of the Fokker 4.0 project who have guided me and contributed a great deal of their time to this research.
I would also like to thank prof. dr. ir. Hans Wortmann who has been a supporting and enthusiastic supervisor. His knowledge and experience about information, information systems and operations was inspiring and helped me in finalizing this work. His feedback was always constructive in nature and guided me in the right direction. Furthermore, I would also like to thank my second supervisor dr. Ying Yang for her feedback and assistance during this project.
Age Mulder
7
1.
Introduction
Manufacturing companies in the aerospace and defence industry produce products of significant complexity for which a large range of information and knowledge is needed (Bishop, 1994; Fransisco Del-Rey-Chamorro & Wallace, 2005). This information has to be of high quality in order to minimize financial and technical uncertainty and to produce more efficiently (Harrington & Blagden, 1999; Harvey & Holdsworth, 2005). In order to preserve and use manufacturing information some sort of documentation is needed (Sundqvist, 2011). This documentation is undergoing large changes within this industry as companies move from a paper to an electronic format, which makes information easier to update and maintain (Del-Rey-Chamorro & Wallace, 2003). However, even highly sophisticated companies find that integrating manufacturing operations into information systems is difficult and struggle with the fact that paper is still shuffled around the offices and factory floors abundantly (Crowder, Hall, Heath, & Wills, 1999).
8 Issues regarding information quality lead to the fact that management and stakeholders make decisions based on information for which no objective quality assurance exist while reliable information is crucial for effective decision-making (Flowerday & Solms, 2005). Documents have to be trusted, but there is no guarantee that they are correct (Sundqvist, 2011). The impairment of information integrity and mistakes made by employees can have costly effects and spread errors throughout the organization (Boritz, 2005). As academia has extensively researched the governance of hardware, software and networks, the focus is now turning towards the governance of information (Tallon et al., 2013). Information governance is found to be a good approach to improve information quality by both researchers and practitioners (Korhonen, Melleri, Hiekkanen, & Helenius, 2013; Otto, 2011). However, the literature in information governance only focuses on organizational structuring and placing decision making authority for data quality management (Weber, Otto, & Osterle, 2009).
9 Only Loshin (2001) wrote elaborately about information ownership and outlines various ownership paradigms. According to Sebastian-Coleman (2013: p.23) “his
approach makes so much sense that it is surprising that organizations are still fighting the battle of ownership”. Despite the recognition of the need for
information governance and information ownership, research on this topic is still scarce. It is not well understood why organizations do not often apply information ownership principles and why the organizations that do apply it continue to struggle with the implementation hereof. Even in heavily regulated industries, e.g. the aerospace and defence, pharmaceutical, semiconductor and chemical industry, which produce complex products and require high quality information (Van Dyk, 2007) the issues and problems of information ownership are not well understood. The implementation of information ownership frequently encounters difficulties in practice (Kropsu-Vehkapera et al., 2009; Redman, 2008). According to Korhonen et al. (2013) empirical research is needed on how organizations organize their data governance and accountabilities.
This research has as purpose to fill this gap in literature by investigating the question why companies continue to struggle with assigning an information owner to information. In order to address these research objectives a case study is executed. Case studies are most often used for explanatory or exploratory research into topics which are not well understood and have the ability to generate answers to ‘why’ questions (Saunders, Lewis, & Thornhill, 2009). This study will try to answer the following research question:
Why do manufacturing companies in the Aerospace and Defence industry struggle with the assignment of ownership to information?
10
2.
Theoretical Background
A distinction between data and information can be made. Data consists of raw facts while information is data that has been given meaning so it has value other than the individual fact (Bellinger, Castro, & Mills, 2004). Information systems use data as input, which is processed to information. Information systems can be seen as “a set of interrelated components that collect (input), manipulate (process),
store, and disseminate (output) data and information and provide a feedback mechanism to meet an objective” (Stair & Reynolds, 2011: p.51). While the
difference between data and information is an important one, in publications on information quality and in practice they are often used interchangeably (Boisot & Canals, 2007; Pipino, Lee, & Wang, 2002; Wang, 1998; Weber et al., 2009). In line with most of these publications, the terms will be used synonymously in this paper unless specified otherwise.
In the following, several concepts related to the ownership issue are further elaborated in order to research why companies are struggling with the implementation of ownership. Since information governance is the overarching theory to ensure information quality, these concepts will be discussed first. Subsequently, information ownership will be reviewed as a mean to assign accountability to information products.
2.1 Information quality
11 A detailed description of what information quality means to the information consumer is given by Wang & Strong (1996), which can be divided into four categories:
1 Intrinsic quality: consists of accuracy, objectivity, credibility and reputation.
2 Contextual quality: consists of value-added, relevancy, timeliness, completeness and the appropriate amount of information.
3 Representational quality: consists of interpretability, ease of understanding, representational consistency and concise representation. 4 Accessibility quality: consists of accessibility and access security.
Whereas product manufacturing processes raw materials into physical products, information processing can be viewed as a small factory that transfers data into information (Loshin, 2001; R. Y. Wang, 1998). As the outcome of a process, the same principles of manufacturing quality improvement can be applied to information quality improvement, of which the most basic one is accountability by the person performing the work (Levis et al., 2007).Similar to product quality, intrinsic quality for information consists of more than accuracy and objectivity that people try to deliver. Information also has to be believable and have a good reputation in order to be considered of high quality. Furthermore, information quality must be considered within the context of the task in order to be perceived as high quality. It has to be easy to understand and interpret. Finally, especially for electronic information, accessibility quality is of importance. This might be less of an issue for paper based information as it does not have to be retrieved from different information systems by the consumers (Wang & Strong, 1996).
12 available at the right place and time, has to be up-to-date and in the correct format (Li & Li, 2000). Among other things, manufacturing information consists of required machine and tool data, estimated cost and time data, the order in which tools and machines are used, bill of materials, engineering designs, technical drawings and the manufacturing conditions (Maeda & Shinohara, 1988; Scallen, 2003; Zachman, 1987). Manufacturing information is all the information necessary for manufacturing a product from conception to delivery (Voss & Winch, 1996).
Loshin (2001) identified various roles in the processing of data, which are responsible for some level of quality assurance:
1. Suppliers: provide information to the system. 2. Acquirers: accept data from external suppliers. 3. Creators: generate data.
4. Processors: take input data and transforms it into output data. 5. Packagers: collect and summarize data for reporting purposes. 6. Delivery agents: deliver packaged information to consumers. 7. Consumers: end user of processed information.
8. Middle managers: responsible for making sure the actors are correctly performing their jobs.
9. Senior managers: responsible for the overall operation of the factory 10. Deciders: senior-level managers who are associated with strategic and
tactical decision-making.
13 consumers. According to Loshin (2001) it is helpful to break down the information flow into a series of processing stages to precisely identify all input and output information streams. This way responsibility can be more precisely associated with a particular actor.
In order to answer the main research question, it must first be understood which roles for quality assurance exists and how information flows through an organization. This leads to the first sub-question of this research:
Which roles, that are responsible for information quality assurance, can be identified at different stages of a manufacturing information flow?
2.2 Information governance
Information governance is defined as “the specification of decision rights and an
accountability framework to encourage desirable behavior in the valuation, creation, storage, use, archival and deletion of information. It includes the processes, roles, standards and metrics that ensure the effective and efficient use of information in enabling an organization to achieve its goals” (Logan, 2010: p.1). It
contains standards, guidelines and controls to guarantee the quality of information within a firm (Kooper, Maes, & Lindgreen, 2011; Lajara & Maçada, 2013). Moreover, it typically includes records management, privacy regulation, information security, data flows and ownership, and data lifecycle management (Kooper et al., 2011). It developed from the idea that information must be governed independently from the underlying technology, should be set free from constraints imposed by compliance and control and must consider intangibles such as interpretation and meaning (Beijer & Kooper, 2010).
14 structure companies have to assign roles and responsibilities to people that are accountable for information (Panian, 2010). When establishing information governance, people, processes and technology should be involved in order to improve the quality of data (Freitas, Michel, Rodrigues, Reis, & Gronovicz, 2013). It can be seen to have two goals: to maximize the value of information by making sure it is reliable, secure and accessible, and to protect information from error, loss of access, inappropriate use or misadventure (Tallon et al., 2013).
Information governance is different from DQM. The governance policy provides the framework, which is comprised of DQM roles. Whereas DQM focuses on the day-to-day activities (Wende, 2007) information governance focuses on long-term data quality improvement rather than error correction. They often propose between three to five DQM roles (Weber et al., 2009). However, the five most common quality roles are not enough to establish an adequate governance model (Korhonen et al., 2013) and there exists ambiguity about the meaning of different roles (Otto, 2011). Loshin (2001) proposes the usage of an information ownership policy as tool to assign all the roles and responsibilities associated with information ownership. However, in essence, this ownership policy is similar to a governance policy. To give an impression of the quality roles within a governance policy a non-exhaustive list of quality roles and definitions is given in Table 1.
15
Quality role Definition Source
Business data
steward Details the corporate-wide data quality standards, rules and policies for a data subject area from a business perspective.
(Wende, 2007) (Dyché & Levy, 2006) (Newman & Logan, 2006)
Chief information officer
Accountable for enterprise information and is responsible for decisions regarding the acquisition, storage, and use of data. The CIO has final say with respect to dispute resolution between areas of ownership and is the ultimate manager of the definition and enforcement of policies.
(Loshin, 2001)
Chief steward Puts the board’s decisions into practice, enforces the adoption of standards, helps establish data quality metrics and targets.
(Wende, 2007) (Dyché & Levy, 2006)
Data
administrator Delivers standardized data element definitions and descriptions in order to both unify and separate departmental views of data with a goal of data integration; may also roll up and maintain enterprise data model based on individual subject area models.
(Dyché & Levy, 2006)
Data steward Manages all aspects of a subset of data with responsibility for integrity, accuracy, and privacy. They have detailed knowledge of the business process and data requirements. At the same time they also have good IT knowledge to be able to translate business requirements into technical requirements.
(Loshin, 2001) (Cheong & Chang, 2007)
(Sebastian-Coleman, 2013)
Data custodian The data custodian manages access to data in accordance with access, security, and usage policies. He or she makes sure that no data consumer makes unauthorized use of accessed data.
(Loshin, 2001)
Data Quality
board Defines the data governance framework for the whole enterprise and controls its implementation.
(Wende, 2007) Data registrar Responsible for cataloging the data sets covered
under the policy as well as the assignment of ownership, the definition of roles, and the determination of responsibilities and assignments of each role.
(Loshin, 2001)
Data trustee Manages information resources internal to the organization and manages relationships with data consumers and data suppliers, both internal and external.
(Loshin, 2001)
Executive
sponsor Provides sponsorship, strategic direction, funding, advocacy and oversight for DQM. (Wende, 2007) Technical data
steward Provides standardized data element definitions and formats, profiles and explains source system details and data flows between systems.
(Wende, 2007) (Dyché & Levy, 2006)
16 Despite the concept of information stewardship being omnipresent, there is no generally accepted definition (Ioannidis, Pym, & Williams, 2014). The lack of a clear definition leads to ambiguity in literature. English (2006: p.2) defines information stewardship as “the willingness to be accountable for a set of business
information for the well-being of the larger organization by operating in service, rather than in control of those around us”, and clearly states that stewardship is
not the same as ownership. Whereas Otto (2011), who also makes a clear distinction between stewardship and ownership, argues that the information owner is accountable for the immediate correctness of information and the steward develops and provides the rules for the handling of data. Information stewards document business requirements, assess the impact thereof and are responsible for the development and provision of the rules for handling the data (Otto, 2011; Wende, 2007). They focus on assuring accuracy, validity, security, management and preservation of information objects (Dawes, 2010) and can supplement the ownership policy (Loshin, 2001).
According to English (2006), ownership implies possessiveness and rights to something while employees often do not own the information they create. They are merely stewards of the information for others who depend on it. However, organizations frequently use the expression ‘to take ownership’ in order to allocate responsibility and accountability for a task or result. This could also be useful for information as it will clarify accountability (Redman, 2008). Sebastian-Coleman (2013: p.21) states that “if it is helpful for organizations to name this
accountability information ownership, then they should”. The information owners
17 Users have historically felt that they were held responsible for the quality of data in information systems that they did not understand. On the contrary, developers find that it is the responsibility of the business user and that they are merely responsible that data is maintained in an accurate and timely manner (Orr, 1998). People, especially those who do no know how the data is produced, often have the assumption that data must be correct when it is stored in a computer (Redman, 1995). While IT is responsible for the information systems, it often does not take responsibility for the information within these systems as they do not want to be held accountable when data does not meet the business needs (Sebastian-Coleman, 2013). They neither create nor use the data (Redman, 1995). Therefore, the business has to take ownership of the data (Sebastian-Coleman, 2013).
2.3 Information ownership
18 ownership as those who “are accountable for the immediate correctness and
consistency of certain data”.
With large amounts of information and many people working with it, rules and guidelines are needed in order to prevent chaos (Stark, 2015). Just like any process needs a process owner who is responsible (Batini, Cappiello, Francalanci, & Maurino, 2009; Stark, 2015) information needs an owner who is accountable for the quality of information (Batini et al., 2009; Loshin, 2001). However, information ownership is often seen as very complex and responsibilities over correcting incorrect data are often not clear. According to Kropsu-Vehkapera et al. (2009), companies implement information ownership in order to prevent and correct product related data errors and to guarantee the quality of information. The desire of identifying an information owner follows as a solution to a problem. When the quality of information is insufficient, inconsistent or chaotic, organizations try to bring it under control by assigning an owner to that information (Sebastian-Coleman, 2013). However, information ownership is often poorly defined and it is not enjoying much success in practice (Redman, 2008). Practicalities of information ownership are complex and often not clear (Kropsu-Vehkapera et al., 2009). It is too often applied as a simple solution to a complex problem. While having a clear accountability defined throughout the information life cycle can be an effective approach, simple solutions do not exist (Sebastian-Coleman, 2013).
19 of data hoarding and the unwillingness of information sharing, as discussed by Levitin & Redman (1998). This solution can lead to resistance, as people are unwilling to give up their ownership rights and people might not feel accountable anymore for the quality of information once they lost their ownership rights (Redman, 2008).
Practical research about the implementation of information ownership is rare and only one author (Loshin, 2001) could be found who provides an extensive description. Sebastian-Coleman (2013) notes that his ownership policy formulates different relationships, clarifies responsibilities, and provides criteria for settling disputes among stakeholders without using the word governance once. While Loshin (2001) recognizes the need for an information owner, much of his work is dedicated to creating an ownership policy. This policy is a tool used to establish all roles and responsibilities associated with information ownership and accountability, as discussed in section 2.2, rather than assigning an owner to information itself. Furthermore, his research identified the paradigms but gives no explanation for when to use them. The only concrete statement about the implementation made is that it depends on the organizational point of view and that different ownership rules may apply in different situations.
The identified ownership paradigms by Loshin (2001: p. 37), which look at who values information and how it relates to ownership, are the following:
1. Creator as owner: the party that creates the data owns the data.
2. Consumer as owner: the party that uses data, and thus cares most about the
value, claims ownership of that data.
3. Compiler as owner: the party that selects information sources and combines
date sets owns the data.
4. Enterprise as owner: the value of information resides with the organization
as a whole and information is owned by the organization as a whole.
5. Funding organization as owner: the user that commissions the data
20 6. Decoder as owner: the party that can ‘unlock’ information owns the
information.
7. Packager as owner: the party that formats information for a particular use
owns the information.
8. Reader as owner: a paradigm that implies that the reader, which adds the
information to a repository, subsumes the value of any data that can be read.
9. The subject as owner: the subject of the data claims ownership of that data. 10. Purchaser/Licenser as owner: the individual or organization that buys or
licenses data takes ownership of that data.
11. Everyone as owner: when data should be available to all with no restrictions
a model of global data ownership exists.
These paradigms can be divided into centralized and decentralized paradigms. In a centralized distribution of ownership a single entity is responsible for ownership of the data, whereas in a decentralized distribution of ownership the roles are separated into various sections of interest (Loshin, 2001). According to Tallon et al. (2013), decentralized users can act as data owners since they have the knowledge to give value to the information. This would most likely correspond with the ‘consumer as owner’ paradigm from Loshin (2001). However, in a case-study among four companies, Kropsu-Vehkapera et al. (2009) found that ownership is often allocated based on business responsibilities, leading to the fact that those who create a piece of information becoming the owner. This possibly leads to an issue if the data consumer is highly reliant on high quality information but he has no stake in the creation of this information. A similar issue occurs when the creator of data does not feel any responsibility about the information in his trust (Loshin, 2001). Literature is inconclusive about the way ownership is assigned.
21 currently try to allocate ownership to information. Therefore the second sub-question of this research is:
Is ownership generally assigned centralized or decentralized, and which paradigms are most common in the assignment of ownership?
2.4 Accountability
In the most agreed-on sense of the word, accountability refers to the ability of being called to render account (Jones, 1992). This is referred to as external accountability, meaning that account is given to someone other than the person being accountable. Internal accountability refers to the personal judgement and adherence to standards irrespective of any external sanctions. Internal accountability has a strong relation with a sense of personal responsibility about the performance of duties (Mulgan, 2000). When applied to information assets, accountability encompasses ultimate responsibility. Every information asset must have a party allocated ultimately accountable for the quality of the asset (ARMA International, 2014; Ladley, 2010). Information quality will not be a priority if nobody is held accountable for ensuring the usability and effectiveness of information. Accountability for information quality can be shared, but it must exist (Ladley, 2010). Information quality accountability ensures that nobody sends poor quality information to other people in the process by making sure that everyone is accountable for the quality of their work (Baskarada, Koronios, & Gao, 2006). According to Khatri & Brown (2010), accountability for information quality can be assigned to either information owners, subject matter experts, information quality managers or information quality analysts.
22 completing an activity (Weber et al., 2009; Wende, 2007). Opposed to normal use, information governance allows for more than one person to be accountable, whereas only one person can be responsible (Weber et al., 2009).
However, in companies it is often not clear who is accountable for making information up-to-date (Kropsu-Vehkapera et al., 2009). Information is maintained within applications but created and changed within business processes (Allen & Cervo, 2015). While databases are static, the world around it keeps changing. In order to maintain information quality a feedback mechanism is needed to make the information in databases represent the real world situation (Orr, 1998). The digitalization of information comes with the risk that many copies of the same document come to exist (Karlson, Smith, & Lee, 2011; Levitin & Redman, 1998). A single document can easily result in multiple files as computers make it easy to copy documents or make various versions of documents across their systems. Subsequently, this can easily result in inconsistency of presumably identical documents. Computers do little to overcome this mess (Karlson et al., 2011). Besides, information is often stored over a multitude of information systems, which makes exchanging information difficult (Hasselbring, 2000). Information is often distributed inefficiently and people, departments and business units hoard data and make and modify their own copies, leading to inconsistencies and conflicts over ownership (Levitin & Redman, 1998). Therefore, it is extremely important for organizations to base a report on a single version of the truth (Khatri & Brown, 2010).
23 When changes occur all relevant documents have to be identified and updated accordingly, which are often dispersed throughout different departments (Pikosz & Malmqvist, 1998).
Therefore the third sub-question of this research is:
How is accountability for the correctness and consistency of information handled within the management for information quality?
2.5 Research setting
The Aerospace & Defence industry was chosen to research information ownership since it recognized the importance of information management early on (Del-Rey-Chamorro & Wallace, 2005). The industry has been subject of numerous information-related studies over the years due to its reliance on knowledge and information (Harrington & Blagden, 1999). Its complex products call for the integration of large amounts of data, information and knowledge created by diverse people (Bishop, 1994; Harvey & Holdsworth, 2005). Therefore, manufacturing execution systems are widely applied within this industry (Chen, He, He, & He, 2010; Van Dyk, 2007; Younus et al., 2010). Furthermore, the industry is characterized by firm compliance standards and regulations that try to maximize safety and airworthiness (Harvey & Holdsworth, 2005).
24 customers, suppliers and authorities. This would make an interesting case to research information ownership.
This research aims to answer the question why companies are struggling with the implementation of information ownership. Three sub-questions are established to help answer this question.
1 Which roles, that are responsible for information quality assurance, can be identified at different stages of a manufacturing information flow?
2 Is ownership generally assigned centralized or decentralized, and which paradigms are most common in the assignment of ownership?
3 How is accountability for the correctness and consistency of information handled within the management for information quality?
25
3.
Methodology
This work has as purpose to research why companies have problems with the assignment of ownership to manufacturing information in the aerospace and defence industry. Research questions aiming to explain ‘how’ or ‘why’ something works often prefer a case study method, especially when contributing to the knowledge on a contemporary topic where the researcher has no control over events (Yin, 2013). However, ‘how’ questions tend to be more of a concern for a survey study (Saunders et al., 2009). Case studies are an approach to explore complex problems within their context (Baxter & Jack, 2008) and allow for in-depth research into the issue (Voss, 2009). Case studies are suitable for exploratory research in topics where the existing knowledge base is poor, ambiguity still exists about concepts and solid theoretical statements cannot be established. It provides a method where people can be interviewed and where events can be observed (Yin, 2013). Furthermore, case studies are the best way to study implementation situations (McCutcheon & Meredith, 1993).
26 For case studies it is important to state what is not included in the research in order to keep a reasonable scope (Baxter & Jack, 2008). Besides, information governance should not be implemented company wide at first, but should focus on the most important aspects initially and to be further developed by continuous improvement (Gartner, 2009). Due to the rapid development of manufacturing informatization and the challenges that still exist in the management of manufacturing information, such as the lack of timely, accurate and consistent information (Liu & Zang, 2014), the unit of analysis of this research is the manufacturing information of the manufacturer. The research explicitly only takes manufacturing information into account, which consists of information that is used to create manufacturing information or information that is used during the manufacturing process (Voss & Winch, 1996).
3.1 Literature study
27
Parameter Preferred Broader
Language English (UK) (US) -
Subject area Information ownership /
Information governance Information management
Business sector Aerospace & Defence /
Manufacturing Regulated industries/ other industries
Geographical area Europe and North America Worldwide
Publication period 2000 - 2015 All time
Literature type Refereed journals and
books Journals, books, articles
Table 2: Parameters for literature study
In order to conduct the search use was made of Google Scholar, EBSCO Business Source Premier and the University of Newcastle library search. Keywords that were mainly used during the search consisted of:
• Information ownership • Information governance • Information management • Knowledge management • Aerospace & Defence
• Manufacturing Information • Information quality
• Accountability • Documentation • Quality roles
In addition to these keywords, several variations have been used which were similar to the listed keywords, e.g. data instead of information and production instead of manufacturing. Advanced search was used in order to combine certain keywords or to exclude certain words (e.g. property rights, intellectual property, proprietary rights).
3.2 Case study protocol
28
Overview of the case study project
The selected case is an aircraft component manufacturer in the Netherlands. The case is selected since it represents a typical case in the industry. Furthermore, the fact that the company recently initiated projects to adopt concepts of Industry 4.0 and smart factory make the case interesting, as they take a new look at the way information is managed. Currently, the organization has a high reliance on paper based information transfer. The main focus of the company’s project is currently on digitalizing the information flow within the manufacturing process. The current paper documentation often flows through the organization with the product. This production order is an accumulation of different documents. The lead-time of a product can be very long and in the meantime information might get out-dated, changed or added. Currently, this is a complicated process with stamps and authorizations. Within the transition to the new situation the company has to establish an information ownership policy that controls who can adjust information objects.
Field procedures
In order to conduct the case study the researcher will be on site during regular work hours. The company has given permission to move freely around the premises, with exclusion of ITAR regulated areas. Furthermore, access is given to a personal computer, with which company databases and information systems can be accessed and a desk is provided in order to be able to work on site permanently. This way access to key information and people could be gained flexibly.
29 and that information will be used anonymously and confidentially. It is clearly stated beforehand that participation is voluntarily and that the interview can be ended at any moment when felt necessary. Interviewees are asked for their permission to record the interview in order to guarantee a correct transcript. While full text transcripts are very common with non-standardised interviews (Saunders et al., 2009), a verbatim record was not established. As proposed by Halcomb & Davidson (2006), a mixed method was applied where field notes are compared to the audio-recording. Immediately after the interview notes and impressions were written down. Finally, the transcript is thematically ordered and reviewed by a secondary party. The results of the interviews have been transcribed into an excel sheet and have been confirmed with the interviewee.
Case study questions
30 Finally, the way that correctness and consistency of information is guaranteed and who is accountable for these actions is research by means of interview questions and a study of internal company documents.
Case study report guide
A descriptive framework for the case study has been established in line with the aforementioned established questions. First the information factory will be described, after which the current owners of information will be identified. Subsequently the reality of correcting information is analysed. The accumulation of these answers will be discussed in the discussion, where the main research question will be answered.
3.3 Data collection
31 Secondly, by means of direct observations the problems were analysed within context. Various tours have been taken through the production process at different times in order to see how and which information is used. The observation of engineers and configuration managers provided an insight into how information is used within the information systems. Thirdly, documentations have been reviewed in order to validate the correctness of previously obtained information and to gain new insights. Fourthly, archival records are accessed for supplementary information. Finally, physical artefacts are studied. These artefacts mostly consisted of printouts from production data. Note that this is an iterative process, when necessary previously used methods were revisited. The different sources of information are one of the strong suits of a case study and has the advantage of developing converging lines of inquiry, a process of triangulation and corroboration (Yin, 2013).
3.4 Analysis of the findings
During the interviews and analysis it was observed that many people have different views on the content of the same document, or use different vocabulary to refer to the same document. To reduce ambiguity, a terms and definition list has been established in order to make sure everyone has the same interpretation about the information objects. An enterprise dictionary fosters good governance since everyone uses the same language. This leads to an increase of the quality of information as it becomes more accurate and reliable (Voskuil, 2015). Data reduction has been applied to the raw interview data in order to create ordered, simplified and easily readable excel sheets.
33
4
Findings
The parts of the company that are of influence on the manufacturing department have been divided into six domains: Human Resources, Manufacturing, Production Engineering, Safety Health Environment & Quality (SHEQ), Supply Chain and Program Management. These domains consist of sub-domains and have been chosen, in accordance with the project management, to create a more manageable overview of the organization. From the six identified domains, five were found to produce information that directly relate to the production process. While Human Resources has an indirect relation with the production process, such as collecting and storing information about operator qualifications, hours and authorizations, it does not directly influence the information used by the manufacturing department. As this paper focuses on the manufacturing information, this department has been excluded from the analysis.
The interviewees were asked questions about which sub-domains they are responsible and which information is needed in order to do their work. Follow-up questions included the location, producer, owner and assessor of this information. As the original interview was conducted in Dutch, a translated example hereof is given in Table 3. In the following sections a descriptive overview is given of the case. The interviews served as basis for the development of this section and were enriched by the study of documents and follow-up meetings.
The main types of information used during manufacturing were: • Engineering data
• Numerical-Control programs • Process instructions (PI) • Production orders (PO) • Planning documents
• First article inspections (FAI) • Technical product
documentations (TPD) • Tooling data
34
Table 3: translated overview of interview
4.1 Quality roles and information flow
35 the specific moulds to be used during the production process. The customers can thus be seen as information suppliers, as they provide the initial information that is needed to manufacture the product.
The company makes all the design data by themselves for turnkey projects. In this case the in-house engineers can be seen as information creators. In addition, external suppliers may provide information about the moulds and machinery, such as milling and cutting machines, and are therefore also information suppliers. All the external information flows through the program management to the production engineering department. The program management functions as information deciders. While they accept data, their main role is to approve the project. The engineering data is often made available through the customers’ systems and, therefore, no actual acquiring takes place. Secondly, configuration management functions as a decoupling point of the information flow for built-to-print projects. They take information from the customers’ system and check the correctness hereof. After they release the information it becomes available for the entire organization. They are the information acquirers, as they take data from external suppliers for provision in the organization.
36 Fourthly, the supply chain domain, consisting of a procurement department and logistics department, creates a week planning, picking orders, in progress work lists, purchasing documentation and production orders. These production orders are the guiding documents for every manufactured article and consists of the routing that determines the route a product will follow through the factory. It contains all information an operator who produces the product needs to know, such as which TPDs and PIs to use and which operation to execute. Currently, this production order is a paper document that travels alongside the product. While production engineering provides most of the information, it is the planning department that decides when a product is going to be taken into production and provides the documentation to the starting production department. They are the packagers and delivery agents of information.
37
38
4.2 Assignment of information ownership
39 However, several issues emerged within the company by assigning ownership based on the creator of the information. By assigning ownership to name, rather than to a role several issues come to exist. It leads to the fact that sometimes the owner of information is relocated to different projects or does not work at the company anymore. Especially during large and new projects there are many people creating new information. Once the project starts to become recurring, there are only few people necessary to maintain the provision of information. The other production engineers then move on to new projects. In these instances the organization does not officially change the owner of the document. This leads to ambiguities when the information appears to be incorrect and when changes have to be made. Moreover, when people make adjustments to information in documents that are not originally created by them, they are technically accountable for the information they add or adjust. In other words, they become the owner of that part of the information, since they created it. However, the owner name does not change, and the repositories are not always sufficient to derive who changed which parts.
40 The other documentations, such as NC-programs, planning and FAIs, do not have an official owner in the sense that there is an owner name written on the document. Nevertheless, similar to the TPDs and PIs, the creator of the file is directly accountable for the correctness. While these files are not always as strictly regulated and do not always have the same iterations of checks by colleagues and members of the quality department, they do follow the same allocation of ownership.
One of the few exceptions on the creator as owner paradigm is when a junior employee creates documentation that has to be checked by a senior employee. This can be the case when a senior production engineer divides the work among junior engineers and functions as mentor. In this case the compiler as owner paradigm is followed as he collects data from multiple engineers that are not accountable for the information they create. However, this is an exception and the only reason the senior engineer takes ownership is because of the lack of knowledge of the junior engineers. While the senior engineer takes accountability for the compiled information, he also helped and guided in the creation hereof. During the interview with the lead production engineer it became obvious that this allocation of ownership might not be the most desired one. The problem that those who use the information have no stake in the creation of the information was recognized and a system where the operator is involved in the creation of the information might be favourable. In this case there would be combined ownership by both the creator and the consumer, but it would be likely that the creator would still be held accountable as he has more knowledge about the establishment of manufacturing instructions.
41 is the task of the configuration manager to make sure the information in the system is up-to-date. This can be troublesome, especially when the customer publishes new information on their site or portal, without any notification. This makes it the responsibility of the configuration manager to check if the information they are working with is always the latest.
A different use for the term ownership was found within the company. Information is ordered within domains called data groups, of which each one has an owner. These people are thus not owners of specific data objects but function more like a gatekeeper. They provide people with access to data domains or refrain people from access. While they are not accountable for the quality of the information and do not interfere with the information, they make sure that only people with the need, knowledge and authorization can reach and modify the documents. Therefore, they can be associated with the middle manager role, which is responsible for some level of quality assurance.
4.3 Guaranteeing the correctness and consistency of information
In the theoretical background it was discussed that ownership is not only difficult to allocate when information is created, but also modified or transformed. During the product development process and production process it must be assured that at all times correct and valid data is used with respect to production documentation. All data used as input for production documentation is under configuration control and is released in the PDM system.
42 As noted by Dyché & Levy (2006), Karlson et al. (2011), Levitin & Redman (1998) and Redman (2008), information is often copied or transformed into other datasets. Also within this case study it was observed that a copious amount of copies are made of information within the company, resulting in difficulties when a change occurs. Figure 2 provides an example, adapted from company documentations, of how engineering data gets copied between different departments. However, all data that is derived from other documentations is traceable to the actual source data that is used at least by indication of type, name and version of this source data.
Figure 2: Example of how information is copied during the information flow
44
5
Discussion
The findings of this research contribute to the scarce literature on information ownership and provide an empirical insight into the problems related to the allocation of ownership. The case study examines why companies struggle with the allocation of ownership to manufacturing information in the Aerospace & Defence industry. The three sub-questions help identify the answer to this problem. Next, the findings will be discussed in context of these sub-questions. Subsequently the resulting implications and findings for the research question will be given.
Sub-question 1: Which roles, that are responsible for information quality
assurance, can be identified at the different stages of a manufacturing information flow.
45 Information creators, suppliers, deciders, acquirers, processors, packagers, delivery agents and consumers all play a role in the realization of information products, depending on the stage of the information flow. In the early stages of the information flow the suppliers and acquirers are key roles that are responsible for some level of quality assurance. Acquirers play an important role as they functioned as an information decoupling point. The information decoupling point represents the point where customer demand data transfers into order data (Mason-Jones & Towill, 1999). Whereas demand data often suffers from delay and distortion, order data is more stable (Swierczek, 2013). The end of the information flow mainly consists of information consumers. The people who create information can be found in all stages. Information flows through many departments that use the information as input for the documentations they create. While they handle the information and are responsible to a certain degree that the information is handled with care, this does not imply that they are accountable. A closer look is needed at the owners of the information products.
Sub-question 2: Is ownership generally assigned centralized or decentralized, and
which paradigms are most common in the assignment of ownership?
46 The creator plays an essential role in guaranteeing the contextual, intrinsic and representational quality of information. The creator can guarantee the completeness, relevancy, interpretability and consistency of the information. Therefore, it might be beneficial to assign the creator as owner when these quality types are of the essence. The creator as owner paradigm as defined by Loshin (2001) states that the party who creates the data is responsible for the immediate correctness thereof. However, it also states that the creator produces the data as a speculative investment as a prelude to recognizing value from the data in the future. In the case company only the former part of the definition was used. The information owner had no intention to create information as a speculative investment.
47 A more centralized concept of ownership was found in with the use of data group owners. While the company calls them data group owners, they do not have any accountability regarding the quality of information within their data group. These owners merely provide or deny access to certain data domains. These data group owners could be seen as data custodians, which are common in information governance structures. Data custodians manage access to data in accordance with access, security and usage policies. They ensure that data consumers do not make unauthorized usage of the data (Loshin, 2001). When accessibility quality is of the essence, these data custodians play a critical role in the provision of access security. Furthermore, the role fulfilled by the quality-engineering department is very similar to the information stewardship role within information governance. Data stewards focus on the development and provision of the rules for handling the data and assure the accuracy and validity (Otto, 2011; Wende, 2007).
Sub-question 3: How is accountability for the correctness and consistency of
manufacturing information handled within the management of information quality?
48
Research question: Why do companies in the Aerospace and Defence industry
struggle with the assignment of ownership to information?
49 Weber et al. (2009) provide a possible solution for the issues of appointing one specific person as information owner. They suggest that more than one person can be held accountable. When having multiple people work with information it is not feasible to only be able to call the person who originally created the document to account. As identified by the first sub-question, many different people handle the information. Similarly, when multiple people work on a new project and the original creators of documents are relocated to a new project, it might be beneficial to have multiple or a new owner. According to Weber et al. (2009), having one role responsible for executing the task guarantees commitment, while having multiple people accountable respects the cooperative culture. However, this will still run into the same problems in the long run, as ownership would still be allocated to specific people. Therefore, this research suggests assigning roles, rather than people, as the owner of information.
A second important finding is that while there is one person officially accountable for the information product, this is not always strictly enforced. Similar to the findings of Kropsu-Vehkapera et al. (2009), information ownership was found to be often allocated based on business responsibilities and thus to be formed automatically. However, if someone modifies a piece of documentation, he is accountable for the modification, even if he is not the owner of the document. Even though there is only one official owner, there can be other people modifying the information. This makes holding somebody accountable for incorrect information extremely difficult, as it might not be the official owner who modified the information product.
50 importance of the information researched. This study limited itself to manufacturing information. Results might be different for more strategic forms of information, created in higher management layers.
51
6
Conclusion
This research found that accountability for information quality is often assigned to the specific person who created information, rather than the role he fulfils. This might not always be the most efficient or best way of managing information quality and it raises certain difficulties that have to be overcome. Companies will have to implement rules and guidelines in order to structure the allocation of ownership. This research suggests the incorporation of information ownership as role in the governance policy. The information owner could be a good addition to the existing roles of information governance, as it is the only role focusing on short-term error correction, rather than prevention.
From a managerial perspective it is essential to realize the importance of information quality. Incorrect information can have severe consequences. It should not be assumed that people automatically take accountability for correcting information. Clear guidelines are needed for the governance of information. Even though information governance policies are not always widely applied by companies, oftentimes they will already have some ways of governing their information. Once clear guidelines and procedures are in place, information ownership can be applied more successfully.
52
References
Adams, J., Khan, H. T. A., Raeside, R., & White, D. (2007). Research Methods for
Graduate Business and Social Science Students. California: Sage Publications,
Inc.
Allen, M., & Cervo, D. (2015). Multi-Domain Master Data Management. Advanced
MDM and Data Governance in Practice. Waltham: Morgan Kaufmann.
ARMA International. (2014). Generally Accepted Recordkeeping Principles.
Baldwin, A. N., Austin, S. ., Hassan, T. M., & Thorpe, A. (1999). Modelling information flow during the conceptual and schematic stages of building design. Construction Management and Economics, 17, 155–167.
Baskarada, S., Koronios, A., & Gao, J. (2006). Towards a Capability Maturity Model for Information Quality Management: A TDQM Approach. In 11th
International Conference on Information Quality (ICIQ-06). Cambridge: SSRN.
Batini, C., Cappiello, C., Francalanci, C., & Maurino, A. (2009). Methodologies for Data Quality Assessment and Improvement. ACM Computing Surveys, 41, 1– 51.
Baxter, P., & Jack, S. (2008). Qualitative Case Study Methodology: Study Design and Implementation for Novice Researchers. The Qualitative Report, 13, 544–559.
Beijer, P., & Kooper, M. (2010). Information Governance: Beyond Risk and Compliance. In Proceedings of the European Conference on Management,
Leadership & Governance (pp. 34 – 39). Poland, Wroclaw: Academic
Publishing Limited.
Bell, J. (2010). Doing Your Research Project (5th ed.). Maidenhead: Open University Press.
Bellinger, G., Castro, D., & Mills, A. (2004). Data, information, knowledge, and wisdom. Retrieved July 26, 2015, from http://www.systems-thinking.org/dikw/dikw.htm
Bishop, A. P. (1994). The Role of Computer-Networks in Aerospace Engineering.
Library Trends, 42, 694–729.
Boisot, M., & Canals, A. (2007). Data , information and knowledge : have we got it right ? In Explorations in information space : knowledge, agents, and
53 Boritz, J. E. (2005). IS practitioners’ views on core concepts of information
integrity. International Journal of Accounting Information Systems, 6, 260– 279.
Brathwaite, K. S. (1983). Resolution of Conflicts in Data Ownership and Sharing in a Corporate Environment. ACM SIGMIS Database, 15, 37 – 42.
Cappiello, C., Francalanci, C., & Pernici, B. (2004). Data quality assessment from the user’s perspective. In Proceedings of the 2004 international workshop on
Information quality in information systems (pp. 68–73).
Chen, J., He, W., He, W., & He, Y. (2010). Research on flexible manufacturing execution system platform. Proceedings - 2010 International Symposium on
Intelligence Information Processing and Trusted Computing, IPTC 2010, 565–
568.
Cheong, L. K., & Chang, V. (2007). The need for data governance: a case study. In
18th Australasian Conference on Informatin Systems (pp. 999 – 1008).
Crowder, R. M., Hall, W., Heath, I., & Wills, G. B. (1999). Integration of Manufacturing Information using Open Hypermedia. Computers in Industry,
38, 31–42.
Dawes, S. S. (2010). Stewardship and usefulness : Policy principles for information-based transparency. Government Information Quarterly, 27, 377–383.
Del-Rey-Chamorro, F., & Wallace, K. (2003). A study of information retrieval in the aerospace domain. 14th International Conference on Engineering Design, 271–272.
Del-Rey-Chamorro, F., & Wallace, K. (2005). Understanding the Search for Information in the Aerospace Domain. In International Conference on
Engineering Design (pp. 1–12). Melbourne.
Dyché, J., & Levy, E. (2006). Customer Data Integration. Reaching a Single Version
of the Truth. New Jersey: John Wiley & Sons.
Eckert, C., Clarkson, J., & Stacey, M. (2001). Information Flow in Engineering Companies: Problems and Their Causes. In International Conference on
Engineering Design (pp. 43 – 50). Glasgow.
Economist Intelligence Unit. (2008). The future of enterprise information
governance. London: The Economist Intelligence Unit Limited.
54 English, L. P. (2006). Information Stewardship : Accountability for Information Quality. In Improving Data Warehouse and Business Information Quality:
Methods for Reducing Costs and Increasing Profits (pp. 344 – 360).
Evernden, R., & Evernden, E. (2003). Third-Generation Information Architecture.
Communications of the ACM, 46, 95–98.
Flowerday, S., & Solms, R. Von. (2005). Real-time information integrity = system integrity + data integrity + continuous assurances. Computers & Security, 24, 604–613.
Freitas, P. A. De, Michel, W. S., Rodrigues, M. A. de M., Reis, E. A. dos, & Gronovicz, M. E. (2013). Information Governance , Big Data and Data Quality. In 16th
International Conference on Computational Science and Engineering (pp.
1142–1143).
Gartner. (2009). Toolkit: Information Governance Project. Retrieved from https://www.gartner.com/doc/933912/toolkit-information-governance-project
Hagmann, J. (2013). Information governance - beyond the buzz. Records
Management Journal, 23, 228–240.
Halcomb, E. J., & Davidson, P. M. (2006). Is verbatim transcription of interview data always necessary? Applied Nursing Research, 19, 38–42.
Harrington, J., & Blagden, J. (1999). The neglected asset: information management in the UK aerospace industry. Business Information Review, 16, 128–136.
Harvey, D. J., & Holdsworth, R. (2005). Knowledge management in the aerospace industry. In International Professional Communication Conference
Proceedings (pp. 237–243).
Hasselbring, W. (2000). Information system integration. Communications of the
ACM, 43, 33–38.
Hightower, R. (2008). Internal controls policies and procedures. John Wiley & Sons.
Hsu, C., & Rattner, L. (1990). Information Modeling for Computerized Manufacturing. IEEE Transactions on Systems, Man and Cybernetics, 20, 758– 776.
Ioannidis, C., Pym, D., & Williams, J. (2014). Sustainability in Information
55 Jarvenpaa, S. L., & Staples, D. S. (2001). Exploring perceptions of organizational ownership of information and expertise. Journal of Management Information
Systems, 18, 151.
Jones, G. W. (1992). The search for local accountability. In Strengthening local
government in the 1990s (pp. 49 – 78). London: Longman.
Karlson, A. K., Smith, G., & Lee, B. (2011). Which Version is This?: Improving the Desktop Experience within a Copy - Aware Computing Ecosystem.
Khatri, V., & Brown, C. V. (2010). Designing Data Governance. Communications of
the ACM, 53, 148 – 152.
Kidd, C., & Burgess, T. F. (2007). Managing Configurations and Data for Effective Project Management. In The Wiley Guide to Project Technology, Supply Chain
& Procurement Management (pp. 108 – 123). New Jersey: John Wiley & Sons.
Kooper, M. N., Maes, R., & Lindgreen, E. E. O. R. (2011). On the governance of information: Introducing a new concept of governance to support the management of information. International Journal of Information
Management, 31, 195–200.
Korhonen, J. J., Melleri, I., Hiekkanen, K., & Helenius, M. (2013). Designing Data Governance Structure : An Organizational Perspective. Journal on Computing,
2, 11–17.
Kropsu-Vehkapera, H., Haapasalo, H., Harkonen, J., & Silvola, R. (2009). Product data management practices in high-tech companies. Industrial Management
& Data Systems, 109, 758 – 774.
Ladley, J. (2010). Making Enterprise Information Management Work for Business:
A Guide to Understanding Information as an Asset. Burlington: Morgan
Kaufmann.
Lajara, T. T., & Maçada, A. C. G. (2013). Information governance framework: The defense manufacturing case study. 19th Americas Conference on Information
Systems, AMCIS 2013, 3, 1984–1993.
Lee, Y. W., Strong, D. M., Kahn, B. K., & Wang, R. Y. (2002). AIMQ: a methodology for information quality assessment. Information & Management, 40, 133– 146.
Levis, M., Helfert, M., & Brady, M. (2007). Information quality management: review of an evolving research area. In Proceedings of the 2007 International
