Quality-Assured Handover of Geographic Data in Building Projects
HANNA GUSTAFSSON
KTH ROYAL INSTITUTE OF TECHNOLOGY
SCHOOL OF ARCHITECTURE AND THE BUILT ENVIRONMENT
i
Preface
This thesis is the final moment of the Degree Programme in Civil Engineer- ing and Urban Management and the Master in Transport and Geoinformation Technology at KTH Royal Institute of Technology in Stockholm, Sweden.
The idea for this study was put forward by Sweco Position AB in Stockholm, where I have also conducted the work. I would like to thank my supervisor at Sweco, David Möller, for your encouragement, feedback, guidance and sup- port. I would also like to thank everyone at the department of IT for the Built Environment at Sweco, Stockholm, for inspiration and nice company.
Further, a big thank you to those who took their time to meet me and partici- pate in the interviews. Without your participation there would be no results of this study.
Finally, I would like to express my gratitude towards my supervisor at KTH, Takeshi Shirabe, and examiner Professor Yifang Ban for valuable feedback.
Hanna Gustafsson Stockholm, Sweden June 2019
ographic data during data collection, and in the building process, but in order for the data to be useful throughout the life cycle of a facility it is important that the data maintains reliability and quality even after the handover from a building project. The handover of geographic data from building projects has, by municipal management units, been described as inefficient. When design- ers do not follow the client’s requirements the data must be handled manually, which is both time consuming and a possible risk for the quality of the data. In a time that place great demands on efficiency, quality and sustainability, it is not justifiable to manually digitizing data that has previously already been dig- ital, furthermore there is also a risk that the human factor and interpretations affect the data quality. There are factors in the handover process that needs to be prevented in order to achieve an unbroken information supply throughout the life cycle of a facility.
In this thesis, a qualitative study has been conducted with the purpose to iden- tify what factors in the handover process prevent quality-assured handover of geographic data from building projects to management organizations. The findings of the study should also result in recommendations of how the City of Stockholm and other municipalities in general can prevent obstacles in the handover process to assure that geographic data maintains its quality through- out the handover. To identify barriers in the handover of geographic data in building projects, the data collection in this study was done through several interviews with project managers, designers and system managers.
The results show that important factors for a successful, quality-assured han- dover of geographic data are related requirements and communication. Final recommendations in order to assure that the geographic data maintains its qual- ity throughout the handover are: early involvement of the management, clear and early communicated requirements, improved communication to achieve a more holistic view, and automated processes that could replace manual work and check data against requirements.
ii
iii
Sammanfattning
Relevant och kvalitetssäkrad geografiska data är kritisk information i flera mo- ment i samhällsbyggnadsprocessen. Det finns olika metoder för att kvalitets- säkra data under datainsamling och under byggprocessen men för att den ska vara användbar under hela anläggningens livstid är det viktigt att datakvalite- ten bibehålls även efter överlämning från ett byggprojekt.
Överlämningen av geografiska data från byggprojekt har av berörda parter be- skrivits som ineffektiv. När projektörer inte följer den kravställning förvalt- ningen ställt måste manuell handpåläggning ske, vilket både är tidskrävande och kan innebära en risk för datakvaliteten. I en tid som ställer stora krav på effektivitet, kvalitet och hållbarhet, är det inte försvarbart att manuellt återska- pa data som i ett tidigare skede redan varit digital. Dessutom finns det också en risk att den mänskliga faktorn påverkar datakvaliteten. Det finns faktorer i överlämningsprocessen som behöver förebyggas för att uppnå ett obrutet in- formationsflöde genom en anläggnings hela livscykel.
I detta arbete har en kvalitativ studie utförts med syfte att identifiera vilka faktorer i överlämningsprocessen som hindrar kvalitetssäkrad överlämning av geografiska data från byggprojekt till förvaltningsorganisation. Studiens re- sultat skulle också resultera i rekommendationer för hur Stockholms stad och övriga kommuner överlag kan förebygga hinder i överlämningsprocessen för att försäkra att geografiska data bibehåller kvalitet under överlämningen. För att identifiera hinder i överlämningen av geografiska data från byggprojekt har datainsamlingen i denna studie utförts genom ett antal intervjuer med projekt- ledare, projektörer och systemförvaltare.
Resultaten visar att viktiga faktorer för en lyckad, kvalitetssäkrad överlämning av geografiska data är relaterade till kravställning och kommunikation. Rekom- mendationer för att säkerställa att kvaliteten hos geografiska data upprätthålls under överlämning till förvaltning är: tidig involvering av förvaltningsorgani- sationen, tydliga och tidigt kommunicerade krav, förbättrad kommunikation, samt att implementera automatiska processer med syfte att ersätta manuellt arbete och kontrollera levererade data mot kravställning.
Thesaurus . . . 1
Glossary . . . 1
1 Introduction 4 1.1 Background . . . 4
1.2 Objectives . . . 6
1.3 Research question . . . 6
1.4 Limitations . . . 6
1.5 Disposition . . . 7
2 Theoretical background and related work 9 2.1 Life cycle management of data . . . 9
2.1.1 Product Life Cycle Management . . . 10
2.1.2 Life cycle management in the built environment sector 10 2.2 Unbroken information flow through digital processes . . . 11
2.3 Previous studies of the handover process . . . 13
3 Methodology 17 3.1 Workflow . . . 17
3.2 Literature Review . . . 18
3.3 Data Collection . . . 18
3.3.1 Qualitative interviews . . . 18
3.3.2 Interviewee Selection . . . 19
3.3.3 Execution . . . 19
3.4 Validity & Reliability . . . 20
3.5 Ethical Considerations . . . 22
4 Results 24 4.1 System managers . . . 24
4.2 Designers . . . 27
4.3 Project managers . . . 29
iv
CONTENTS v
4.4 Summary . . . 31
5 Analysis & Discussion 33 5.1 Communication . . . 35
5.2 Requirements . . . 36
5.3 Digitalization & Technology . . . 37
5.4 Case - Traffic Administration Office . . . 37
5.5 The handover process . . . 39
5.6 Data collection . . . 39
6 Conclusions & Future Work 42 6.1 Conclusions . . . 42
6.2 Recommendations . . . 43
6.3 Future work . . . 44
Bibliography 46
A Interview guide - Project 50
B Interview guide - Management organization 52
CONTENTS 1
Thesaurus
Exploateringskontoret - Stockholm City Development Administration Stadsbyggnadskontoret - The Urban Planning Department in Stockholm Stockholms stad - The City of Stockholm (Municipality of Stockholm) Teknisk handbok - Technical manual
Trafikkontoret - The Traffic Administration Office in Stockholm
Definitions
Automation - The technology by which a process or procedure is performed with minimal human assistance.
BIM - Building Information Modelling.
Building process - Includes the planning, construction and management of cities, buildings, infrastructure, facilities and the cultural and natural environ- ment.
Built environment sector - Includes companies with operations in the con- struction industry, property management and building materials industries, as well as architectural firms and technical consulting companies as well as pub- lic actors, such as authorities, municipalities and county councils, who both act in the sector through their role as builders and are also responsible for plan- ning processes, rules and legal requirements.
Digitization - The process of changing from analogue to digital form.
Digitalization - The use of digital technologies to change a business model and provide new revenue and value-producing opportunities.
DWG file - An AutoCAD Drawing Database file, storing two- and three- di- mensional design data and metadata.
Geographic data - Also known as geospatial data or geodata. Data char- acterized by a certain location on earth, expressed by means of geographical coordinates.
GIS - Geographic Information System. A framework designed to capture, store, managing and analyzing data.
Handover process - The process where data is delivered from the project or- ganization to the management organization.
Life cycle management - The process of managing the entire life cycle of a product.
Management phase - An organization’s coordinated activity to store its as- sets.
PDF - Portable Document Format. An open standard for exchange of elec- tronic documents.
Requirements - Technical requirements for management data. Description of how the management organization/client wishes to receive the data, includes technical requirements such as format, coordinate system, attributes etc.
2D - Two dimensional.
3D - Three dimensional.
3
Introduction
The introduction will give a brief explanation of the problem analyzed and its background. The section presents the problem formulation, purpose and ob- jective, research question, limitations, and finally the disposition of the report.
1.1 Background
The built environment sector is Sweden’s largest single sector, according to the Swedish Construction Federation (Sveriges Byggindustrier) it accounts for about half of the national wealth of Sweden [1]. The building process con- sist of several phases - planning, design, construction and management - with information, including geographic data, being shared between them. It is a fragmented sector with many stakeholders and processes, and with stakehold- ers operating inside their traditional comfort zone a holistic approach is often lacking.
Geographic data is critical information in the building process where access to relevant and up-to-date geographic data is a prerequisite for an efficient con- struction and management process as well as for communication with decision makers and the public. Non-quality-assured data means a risk, which makes handover and management of geographic data of great importance. There are many methods for quality-assuring geographic data in data collection, and in the building process, but in order for data to be useful throughout the life cy- cle of a facility it is important that data maintains reliability and good quality even after being handed over from a project. Previous research on engineering projects has examined coordination across the groups involved in the handover process [2], noting the short-term focus on the project as opposed to the long-
4
CHAPTER 1. INTRODUCTION 5
term view of the output of the project [3]. It is stated by Almqvist et al. [4], among others, that there must be more focus on the handover process and de- liveries of data.
From building and construction projects geographic data is handed over to the management organization in as-built documents, which are drawings that have been modified based on deviations from the original drawings that were made during the construction phase. In the City of Stockholm, recipients of information and data from urban construction projects are mainly the various units at the Traffic Administration Office, who will further manage the ob- jects. How technical information is to be designed for delivery is described by requirements in the technical manual of the Traffic Administration Office in Stockholm (Teknisk handbok), that is aimed at consultants and other external actors in a building project. For different plan assets, there are various admin- istrative databases.
The handover of geographic data from building and construction projects has, by municipal management units, been described as inefficient and risky for the data quality and reliability. The problem originates from designers who do not deliver data that complies with the requirements. When the delivered data does not fulfill the requirements, there can occur problems when convert- ing the data into the GIS-database. As a case study, and to delimit the work, the study in this report will focus on the Traffic Administrations Office’s man- agement of park data.
Although there are requirements in the technical manual for geographic data that should be delivered to the database of parks, the data that the system man- agers of the Traffic Administration Office receives is in ten of ten cases is in- correct, and time-consuming work is required to interpret the drawings and manually update the database. This is a major risk for the quality and reliabil- ity of the geographic data, as the system manager must manually interpret the drawings and recreate the objects in the database.
In a time that place great demands on efficiency, quality and sustainability, it is not justifiable to manually handle data and recreate information instead of using what was created in earlier stages. Digitizing data that has already been digital in an earlier phase is inefficient and it is a risk that the human factor af- fects the data quality. In the built environment sector there are great visions of unbroken information flows and a completely digital building process [5, 6, 4,
7, 8] but if such a basic thing as the handover of two-dimensional geographic data from park projects does not work, there are factors in the handover pro- cess that needs to be prevented in order to achieve an unbroken information supply throughout the life cycle of a facility.
1.2 Objectives
The objectives of the study are to identify barriers in the handover of geo- graphic data from project to management in order to preserve the quality of the data in the handover, and to provide recommendations to achieve an un- broken information supply throughout the whole life cycle of the facility.
The findings of the study should result in recommendations of how the City of Stockholm and other municipalities in general can prevent the obstacles in the handover process to assure that the geographic data maintains its quality throughout the handover.
1.3 Research question
What factors in the handover process prevent quality-assured handover of ge- ographic data from building projects to management organizations?
1.4 Limitations
In order to investigate the issue to a satisfactory level within the set time limit following delimitations have been made.
In this study the term data refers to geographic data. A building project ad- dresses many issues. Some are spatial or , but others may be financial or ad- ministrative and requires data of the corresponding types. In the handover from a project to the management organization, many different types of data can thus be included. However, the aim of this study deal with technical doc- umentation and geographic issues and data such as drawings, models, geo- graphic data and databases, not management information such as management plans, inspection protocols and checklists.
Further, with respect to requirement documentation and control of the require- ments the focus in this study is not on assessing the requirements and investi-
CHAPTER 1. INTRODUCTION 7
gating how the requirement setting should be designed. Also, the study does not aim to investigate the actual transformation between different data formats, such as DWG and shapefiles, since that is not the issue, and such solutions do already exist.
1.5 Disposition
The report has been outlined according to this section, and the purpose is here to give the reader an overview of how the report is structured.
• Introduction: Gives a brief explanation of the problem analyzed and its background. Include problem formulation, purpose and objective, and limitations.
• Literature review: Gives a theoretical context to the study and cover previous research. This to ensure understanding of the subject and to help conclude findings to the research question.
• Methodology: The methodology explains how the research in this report has been conducted and how the different parts of the research has been structured.
• Results: The results from the empirical study are presented.
• Analysis & Discussion: Analysis based on the results in regard to the literature and theory, and discussion around the analysis.
• Conclusions: Conclusions are presented based on the results and the discussion. The research question is answered, and future studies and recommendations are presented.
Chapter 2
Theoretical background and re- lated work
2.1 Life cycle management of data
The term life cycle may refer to different things depending on the area of in- terest. In terms of data management, the term is often expressed as data life cycle or information life cycle, which refers to the process of planning, collect- ing and storing data. USGS [9], United States Geological Survey, illustrates the stages of data management and describes the data flow through a research project in the USGS Science Data Lifecycle Model (Figure 2.1).
Figure 2.1: USGS Science Data Lifecycle Model. (Source: [9])
In product management the term life cycle is referred to as product life cycle (PLC), and life cycle data is data that supports the production process includ- ing design, construction and maintenance of a product and that is transferred between the phases. This way of modelling life cycle data is also becoming
9
more common in the area of Building information modelling (BIM), which implies that it affects geographic data when BIM and geodata is integrated.
2.1.1 Product Life Cycle Management
According to Kiritsis et al [10] the product life cycle can be defined by three main phases. Beginning-of-life including design and manufacturing, middle- of-life including distribution, use and support, and end-of-life where products are retired in order to be recycled or disposed. Today PLM primarily focus on the beginning phase of the product, almost no product information is trans- ferred to the middle and end phases, and Terzi et al. [11] see an upcoming need for PLM also in these phases. Such PLM, including all phases of the product life cycle, would give feedback from clients to designers and provide service, maintenance and recycle operators, independent of their geographic location, with product information that is up to date.
A prerequisite for a sustainable product life cycle management (SPLM) is the ability to being able to handle product information in the later phases of the product life cycle, as its information after it has reached the client is needed in order to recycle or dispose a product in an effective way. Its importance is stated by Vadoudi et al [12] , and since territorial aspects are also important from an environmental point of view the authors propose a combination of SPLM and GIS to improve the SPLM.
2.1.2 Life cycle management in the built environment sector
To overcome the issues of an ineffective construction life cycle in the built environment sector a holistic vision called Integrated Design Delivery Solu- tions (IDDS) has been taken by the International Council for Research and Innovation in Building and Construction [13]. The use of IDDS involves changes in all stages of the supply chain, from planning to design, construc- tion, commissioning, operation retrofit and decommissioning. In order to achieve holistic views of construction by fulfilling the idea of adding project and whole-life value in all phases and for all stakeholders the evolution in- cludes involved collaborative processes, integrated information and automa- tion systems, enhanced skills and knowledge management. In more detail, Prins and Owen [13] means that by analyzing the processes for each phase throughout the life cycle of the project, the effects of the processes will be
CHAPTER 2. THEORETICAL BACKGROUND AND RELATED WORK 11
improved which could enable more effective processes. For a seamless in- formation flow between different phases technologies for collaboration and automation are also important, and for this, skills about advanced information technology will be needed.
2.2 Unbroken information flow through dig- ital processes
A factor that affects the building process, and which is a current topic, is the digitalization. In 2016 the innovation program Smart built environment (SBE) was established. The program aims to, through efficient information management, industrial processes and increased digitalization, lead to sus- tainable building and maximum user benefits. To reach an unbroken informa- tion flow and take advantage of all possibilities of digitalization, integration of BIM, GIS and industrial processes is applied [5]. Within the program several projects have been, and are currently, implemented with the aim of digitizing processes for a more efficient uninterrupted flow of information over the entire life cycle of a facility.
Argus et al. [6] have carried out a project with the objective to integrate BIM models with geographic data for a simpler and more flexible information flow in the building process, since today’s process lacks continuous communication and information flow between different authorities. This results in duplication of work and long processing times. The background of the project was in- vestigated by Almqvist et al. [4] where it was examined what conditions that exist for an effective information exchange of 3D data, what development po- tential exists, and possible applications areas. In the preliminary investigation the conclusion could be drawn that for an effective handover requirements for the data to be delivered not only need to exist - the requirements must also be followed. Almqvist et al. [4] also highlights the fact that all the players in the industry need to have a holistic perspective, with knowledge and under- standing of each other’s work and sub-processes rather than only focusing on their own responsibility and budget. This to create information with regard to usability in the next step of a building project. It is stated that information supply must be streamlined with focus on handovers and deliveries.
In order to meet the wish that information that is worked out in the different phases of the process should be able to be used smoothly by the next actor in the next stage of the process the information needs to be digital and standard- ized [8]. The strategic project DigSam [8] aims to create an unbroken, digital, standardized information flow, particularly with focus on the early planning stages of the process - to carry out comprehensive planning and citizen dia- logue using the technical possibilities of digitization. Furthermore, the project aims to map out the possibilities for easier storage and access of data, and what perceived juridical obstacles exists for a digital building process.
Based on the challenges that arise in the tension between the building indus- try’s special features and the conditions for a successful digitization-driven industry development Jacobsson et al. [7] aims to create understanding of how the parties in the industry experience and manage these challenges in their work to develop and streamline the building process. The authors do also describe and analyze how the obstacles can be overcome. Through qualita- tive interviews and focus groups with different industry players - facility man- agers, contractors, consultants and architects - obstacles and motivations for a digitization-driven industry development were identified. Identified barri- ers are conflicting roles and perspectives, routine networks and identities, as well as lack of pressure to change. The identified motivations are pockets- of-change, technology maturity and infrastructure base. Within the project the participants have also worked on a number of scientific contributions that from different perspectives analyze and highlight the overall purpose of the study. Linderoth et al. [14] have investigated the issue of digital transforma- tion in relation to the characteristics of the building and construction industry.
The purpose of the study is to explore how the interaction among clients, con- tractors, consultants and architects shape industry characteristics and options for digital transformation through adoption and use of digital technology. The conclusion is that it is the focus on practical day-to-day action, the heteroge- neous client side who have difficulties in putting demands on contractors and subcontractors due to lack of competence, and a product that raise barriers for process innovations, that constrain digital transformation.
CHAPTER 2. THEORETICAL BACKGROUND AND RELATED WORK 13
2.3 Previous studies of the handover process
Schneider et al. [15] performed a study initiated by the municipality of Trond- heim of the handover process of commercial buildings. The authors analyze the causes of building delays and defects identified in the process along with countermeasures to re mediate these. Norway has a standard which clearly describes the handover of buildings and civil engineering work, but the pro- cess is still too costly, time consuming and inefficient. The literature review conducted by the authors shows that the handover process presents a challenge internationally, but little research covers the issue. From interviews conducted it was found that the main causes of handover challenges can be divided into three areas. Most of the process challenges were related to complex contracts and vague specification, which cause differing expectations and interpreta- tions from client and contractor. A more detailed specification of work in the contract would leave less room for interpretation. Other causes of handover challenges according to Schneider et al. [15] are poor execution and planning of handover related activities, and challenges related to damaged data, often caused by late fixes.
From fieldwork conducted ahead of the 2012 Summer Olympics in London Whyte et al. [16] develop understanding and builds a theory of how organiza- tions hand over digital data from project to management. The analytic work focused on understanding the practices involved in collating and handing over the digital data, and how knowledge was coordinated across the separated groups involved. Professionals in data handover interviewed by the authors highlighted a detailed plan of data requirements as necessary for a successful result, since it is the failure to completely specify the requirements specified at the outset that leads to a substantial knowledge coordination between project delivery teams and operators in the final stages of the project. Winch [17] ar- gues that contribute with good and clear requirements for the data to be handed over can be difficult for the client as the client hires a contractor who have better knowledge of what is ordered than the client himself has, a so called principal- agent problem. The entrepreneur’s interests do not necessarily have to be the same as the client’s, which means that the client cannot expect the contractor to act according to the client’s needs.
Anthony [18] capture lessons learned and success factors from projects that have had more or less successful handovers. Identified pitfalls and good prac- tice are distilled into guidance for future projects, to mitigate the risk of poor
handovers and improve the likelihood of successful project handovers. The result of the survey and interviews conducted by Anthony [18] is that the re- quirements should be written into the contract in as much detail and as specif- ically as possible, since the more accurate specifications the more likely it is that the output fulfill the needs. The requirements should also be agreed on at the outset of the project to ensure that is it clear for all parties of the project.
Furthermore, the handover should be seen as a process, not a date, and should hence be planned for from the start of the project with end users involved from the outset.
Through interviews with leading clients and their supply chain Whyte et al. [19]
also gives recommendations for data handover from civil engineering projects to management of built infrastructure. The handover of data from civil en- gineering projects to management was found to often be poorly implemented, with key engineers leaving the project before the data handover. Another chal- lenge faced was to ensure accuracy and completeness of the delivered data, as the data is not checked thoroughly before delivery. At the same time he facility management rely on that the right information has been provided and has be- ing checked. For continuity across the life-cycle interoperable standards and systems are seen as crucial, since errors in data are often introduced in transla- tion. Based on the findings, Whyte et al [19] formulated recommendations for actions at project level and actions at corporate level. Actions at project level included defining data requirements at the start of the project to ensure that ac- curate, complete and relevant data is delivered in the most appropriate format in the end of the project. Also, it is recommended to start handover processes early to get facility managers engaged before key engineers leave the project, which allow shared learning and possibility to go through the requirements.
Actions at corporate level include industry initiatives to develop interoperable systems and standards that support the data exchange requirement.
Mäkelä [20] has carried out a qualitative case study of handover in mega projects where key factors that lead to success and failure throughout the han- dover are investigated. Through literature and empirical studies six factors for a successful handover are found: foresight, good communication, clear roles, committed leaders, committed operators and forms of partnering. With fore- sight Mäkelä [20] mean time to improve the handover performance by invit- ing participants in the handover work and capture their views and suggestions.
Good communication improves cooperation and could avoid errors, it does also improve understanding of others roles and work in the handover. Com-
CHAPTER 2. THEORETICAL BACKGROUND AND RELATED WORK 15
mitted leaders improve projects which improve the preparation of the handover and committed operators could avoid unexpected deliveries and make them more satisfied. Finally, forms of partnering and economic incentives can im- prove the performance of handover in mega projects and avoid optimism bias.
In the study four failing factors were also identified: Strategic and optimism bias, lack of time and lack of communication.
Chapter 3
Methodology
The methodology section explains how the research in this report has been conducted and how the different parts of the research have been structured.
The methodology includes the research design of the report and how the data collection and analysis have been conducted. A review of the validity of the work as well as ethical aspects of the report are also included.
3.1 Workflow
The workflow that has been followed is described schematically in Figure 3.1.
Figure 3.1: Workflow described schematically.
17
3.2 Literature Review
A literature review covering previous research has been done to give a theoret- ical context to the work, and to help conclude findings to the research question.
The literature is important to relate findings in the qualitative data back to pre- vious research.
The literature review has been performed by reviewing research articles, re- ports, books and web pages. In order to find sources, searches in databases such as KTH PRIMO’s library, Google Scholar and Diva have been made.
3.3 Data Collection
The research question What factors in the handover process prevent quality- assured handover of geographic data from building projects to management organizations? requires understanding for both sides of the handover. To answer the research question the collection of data has been done through interviews. The interviews were conducted with representatives from three main groups of people involved in the handover of geographic data: design- ers, project managers and system managers. This intended to cover different points of view of the handover, and to understand what obstacles those who design, those responsible for the project and those who receive and should manage the geographic data experience.
3.3.1 Qualitative interviews
Qualitative studies aim to that from people’s experiences and perspectives be- ing able to describe, explain and interpret an issue. In qualitative studies, it is assumed that reality can be perceived in many ways, and consequently there is no absolute and objective truth [21]. The studies often include a small number of people, but the experiences of these people are investigated more deeply [22]. The research question in this study requires understanding and opinions of those involved in the handover process, which makes the method of qualitative interviews suitable.
CHAPTER 3. METHODOLOGY 19
3.3.2 Interviewee Selection
Interviewees were selected with the intent of collecting relevant information and understand the issue from different points of view. Three system managers from the Traffic Administration Office in Stockholm were asked to participate because they would have the possibility to answer questions from the manag- ing organization’s point of view. Furthermore, six people, project managers and designers were asked to participate to include the perspective of the other side of the handover. To investigate the issue of quality-assured handovers of geographic data both in general and specific regarding the Traffic Administra- tion Office’s management of park data, only designers and project managers who have previously delivered geographic data to the park database were in- terviewed. Out of the people asked to participate, six were finally interviewed.
The three who did not attend declined or got prevented too late to be replaced.
Despite that the sample was smaller than desired, the purpose to cover differ- ent views and experiences of handover of data was considered fulfilled.
To avoid focusing on gender, age or background, all respondents remain anony- mous in this report. In order to still be able to distinguish opinions and high- light who said what, each interviewee is represented by a title, e.g "Designer A". A summary of the interviewees is presented in Figure 2.3.
Figure 3.2: Presentation of interviewees.
3.3.3 Execution
In order to retrieve as much information as possible from the interviewees by letting them talk freely in the conversation the interviews were semi-structured.
A low degree of structure gives the interviewee more room to interpret the questions based on own experiences, and a low degree of standardization makes it possible to customize the planned questions to the conversation. Low struc- ture and low standardization make it possible to make a qualitative analysis of
the results [23].
In semi-structured interviews, an interview guide is created that serves as a template and guidance through the interview. The interview guide should con- sist of several themes with related questions, but the questions can be formu- lated differently on different occasions and the order of the questions can be different at different interview sessions [24]. During the interviews interview guides with general, open ended questions were used. Two different interview guides were used depending on the interviewee, since the questions were for- mulated differently depending on whether the interviewee is working in the project or in management. The interview guides can be found in Appendix A and Appendix B. Since both the interviewer and the interviewees of this study were Swedish speaking, the interviews were held in Swedish with the intention to avoid misunderstandings due to the language and to give the in- terviewees the possibility to express themselves as free and natural as possible.
Each interview was conducted at the respondents’ place of work to minimize the time spent and facilitate the opportunity to find a time to perform the in- terview. For each interview one hour was booked, and most interviews lasted between thirty minutes and one hour. To prevent data from being lost all in- terviews were recorded. According to Trost [25] recording the interviews also contributes to a good result as it allows the interviewer to concentrate on the conversation and the dynamics of the interview instead of taking notes.
After each interview had been carried out it was transcribed. The transcription is a very time-consuming task, and cause challenges described by Greener [26].
On the other hand, the transcription of the recorded material from the inter- views led to a much more straight forward process of analyzing and were con- sidered as very valuable in the analysis since results from multiple interviews could be analyzed at the same time.
3.4 Validity & Reliability
When writing a report, a reflection of the validity and reliability of the work must always be done. Validity measures the extent to which what is to be in- vestigated is really investigated, and reliability refers to that the examinations and measurements are correctly performed [27]. The validity and reliability are shown in many ways, but the focus always lies in using a method that makes it possible to recreate the results. According to Collis and Hussey [28] a high
CHAPTER 3. METHODOLOGY 21
level of reliability and validity has been achieved if the results from the study are likely to be replicated with a similar outcome and if the conclusions drawn from the research can stand up against critical review.
To strengthen the validity of this study and the interviewees importance in delivering valid results the method of the interviewee selection and their rele- vance has been described. Regarding the reliability of the report it is possible that the result can have been affected by the author having possible difficul- ties in being completely objective in the process and analysis of the research, which the reader should be aware of this and take it into consideration when reading the report.
A factor that directly affects the results of this study is the selection of in- terviewees. If other professionals would have been interviewed, the answers would probably also had differed from the collected materials in this study. By including professionals involved on both sides of the handover process differ- ent points of view are given, which is considered to increase both the validity and the reliability of the study. The accuracy of using a qualitative method can also be influenced by how the author transcribe the material. In this study this is considered to have been met by all the interviews being listened to several times during the execution of the transcription. The study’s validity could also have been increased through more interviews, which, however, was limited by the time aspect of the study.
The methodology of qualitative research is frequently criticized for lacking scientific rigour with poor justification of the methods adopted, lack of trans- parency in the analytic procedures and the findings only being a collection of personal opinions subject to researcher bias [29]. Collis and Hussey [28]
mention that the importance of reliability in an interpretive study might not be as crucial as to other types of studies, as the case study works under cir- cumstances that might be hard to replicate. However, the study still requires a level of reliability, but the focus might instead lie on increasing the ability to generalize the results, and increasing the validity of the study.
3.5 Ethical Considerations
When interviews are used as research method and there are subjects who have been interviewed, there are ethical considerations to process and follow [26].
During the interviews it was made sure that the purpose and aim of the study was clear for everyone involved. Everyone that participated in the interviews was also notified that everything recorded would only be used for this study, and it was made sure that everyone gave their consent.
23
Results
As the collection of data has been done through interviews, the data that the result and conclusions of this study are based on is hence answers from the interviews. The results will be presented in terms of opinions and experiences of the different professions. For each section, the results are first summarized and presented in a table, followed by a more detailed presentation of the results.
Finally, in the last section a summary of the results can be found.
4.1 System managers
Figure 4.1: Summary of results.
24
CHAPTER 4. RESULTS 25
The most substantial problem that System manager E and F experiences from the handover is that data that is delivered does not fulfill the requirements and may in some cases be completely unusable. The deliverable usually consists of either only pdf files, or DWG files which lacks structure. For the system managers it is both time consuming to have to search in the files for what is needed, but even more time consuming it is to manually have to interpret the drawings and re-create the new objects in the database. Except from the time aspect it is also a question of data quality and accuracy since it will always be a risk that the data is affected by individual interpretations. Interpreting and manually creating data in the database also a great risk of data being lost, and the accuracy of the data being reduced as the drawings delivered are of- ten unclear and difficult to interpret. The reliability of the data relies on the individual and the background map used, which gives great uncertainty (Sys- tem manager E). As a solution to this System manager E desires some type of compulsory process, where project managers and designers are forced to submit correct files. A wish is also expressed to automate processes that to- day must be performed manually, such as reviewing and sorting information.
For the system managers there is a lot of unnecessary work after the handover, which could be made much more efficient (System manager F). Overall, the system managers experience that the management and the problems that the management organization are facing are given too little attention in relation to the actual distribution of a facility’s lifetime. It is perceived to be far more fo- cus on the earlier stages of a building project, even though the actual building process only represents a fraction of the total life cycle of a facility.
An obstacle mentioned is that the requirements are often communicated late in a project. For example, the requirements for deliveries to the park database are found in a late phase of the technical manual, which assumes that the entire manual is read through at an early stage (System manager E and F). System manager F finds, though, that when the requirements are communicated ear- lier the people involved in the project are very accommodating. The fact that requirements are not followed can partly be explained by the fact that, as pre- viously mentioned, they are communicated too late in the project or that the management has not been clear enough and made demands, but also that those who produce the data have their own routines and templates that they would rather follow (System manager E).
System manager E and F sees it as a major disadvantage that they are not involved in the project until the delivery, and emphasize it as a strong wish to be involved earlier in the process in more projects as the outcome when it has happened is that they have received the data they wanted. These occasions have only taken place in larger projects but is something that is considered desirable even in smaller projects. It can be problematic to reach the one that should be fed back to regarding incorrect files. This since the system manager only has contact with the project manager, not the one who established the drawing and created the data (System manager E). It also appears that despite the extra work that explicitly emerge because of incorrect files being delivered, no feedback is given to the project managers, with the motivation that the sys- tem managers are happy if they receive any data at all (System manager E).
Regarding the organization in City of Stockholm and especially within the Traffic Administration Office the system managers highlights internal deficien- cies. There are uncertainties about which objects and facilities that have been handed over to the management, and which that the project is still responsi- ble for. Problems have also occurred due to unclear communication regarding where projects are going on (System manager E). This is problematic since the management does not know from where data is expected to be delivered, and thus does not know when data is missing (System manager F). Further- more, there are no uniform routines and working methods for managing and updating information. In addition, there are also uncertainties in whether the information is up-to-date and accurate, and sometimes information is missing (System manager E).
As mentioned, The Traffic Administration Office in Stockholm and the city in general lack a common information management function for receiving, or handing over, geographic data and other digital information deliveries between project and management. Instead, the projects must communicate with each of the management organizations, which all have their own requirements (Sys- tem manager E). System manager F’s experience is that the projects want to deliver what that the management demands and can use, but it is difficult when the municipality itself does not have a mutual plan, as requirements and rou- tines vary depending on which unit the data is to be handed over to. In the organization of City of Stockholm various management units within the same municipality have different management systems and databases and are thus interested in different data, and also there are different requirements for the
CHAPTER 4. RESULTS 27
geographic data depending on the management organization (system manager F). System manager E express a wish that the different administrations should work more uniform and together, instead of having different systems, working methods, routines and data requirements.
4.2 Designers
Figure 4.2: Summary of results.
In the beginning of many construction and civil engineering projects there are deficiencies from the clients themselves in the requirement setting of the ge- ographic data that should be delivered. According to Designer B this is due to that those who wrote the documents with requirements lack competence in the area of subject. If there exist an ambiguity and different opinions about the requirements there is a risk for the "order-consultant frustration", where both parts feel misunderstood (Designer A). If the scope of the extra work that the deliveries require is not understood in the early stages of a project the calcula- tions are made from a standard procedure (Designer A), which is a cost issue as costs increase with late changes, and the extra work is often not something that is included in the budget from start (Designer B). According to Designer A it is of great importance that the details of the final result of the project are clarified in the early stages of the project in order to be able to carry it out and hand it over with as good results as possible. In order to avoid a time and cost
issue later in the project, it is important that the requirements and demands are made clear in an early stage (Designer A).
The specification of requirements can vary greatly depending on the project.
In cases where there are no requirements for the geographic data, a reasonable estimate is made based on how it usually is in similar projects (Designer B).
This can complicate the work and become a costly and time-consuming pro- cess since requirements and templates are not consistent and vary depending on where the data is to be delivered (Designer A). Most faults usually occur when the requirements that are set differ from the usual working method, and lack of standards is something that all respondents experience as problematic.
Designer A explains that there are different "schools" in different companies which makes people from different companies interpret differently. If there are no clear requirements it is hence a risk that the result will differ dependent of who conducted the work. In order to avoid this Designer A and B express that the document with requirements should be clear and detailed and be given very early in the project.
Designer A and B claims that they use to be good at performing self-checks based on what the client ordered and what is to be delivered. One problem De- signer A sees, however, is that the controls often take place in long stages with many people involved, which can lead to information being changed along the way. The large distance between the management/client and the designer can also cause problems when things that are very important for the management does not feel as important for the designer and are consequently forgotten or ignored (Designer A). Despite early and clear demands, the human factor is al- ways a risk as the requirements can be lost or forgotten by the designers during the work in the project (Designer B). An important aspect to avoid the risk that the requirements are lost or forgotten is that the project manager is responsible for conveying the requirements, and also to ensure that they are followed and incorporated. This is especially important in cases where the recipient of the data is new for those who are part of the project (Designer B).
A large amount of the information worked out by the designers is not used later on in the process. Especially when more advanced BIM modelling is used instead of 2D drawings in larger projects (Designer A and B). With to- day’s technology and knowledge, it would be possible to deliver something much better and more useful, but in most cases there is no competence or management system that has the capacity to receive that kind of data, and then
CHAPTER 4. RESULTS 29
much information is lost (Designer A).
To avoid problems due to the large distance between the management/final client and the designer causing would according to Designer B be meetings between those involved. Designer B has positive experience from such meet- ings, as it gave new information and perspectives and in dialogue about work- ing methods it was realized what opportunities that exist for the delivery (De- signer B). From own experience of the ordering side, Designer A confirms the benefits of having contact between the designer and the manager, to be able to adapt the result and to be able to understand the problems of both sides. A certain requirement might mean a giant amount of work from the designer’s point of view, but in the same way another requirement might be imperatively necessary for the management to be able to receive the data and further use it.
4.3 Project managers
Figure 4.3: Summary of results.
The project managers interviewed, Project manager C and D, do not think that they have the competence to control that the deliveries satisfy the requirements and the clients demands, but rely on that the designers have checked it and that the management will check it.
There is no structure and management-specific routines to ensure that the in- formation supply of the Traffic Administration Office in Stockholm and the Stockholm City development administration is efficient and that the informa- tion will be available in the future. Although the entities work relatively well individually, there is a lack of connection between them (Project manager D).
The lack of feedback mentioned by System manager E is confirmed by the project managers, who claims that there have never been any problems in the delivery to the management (Project manager C and D).
What is mentioned as the largest obstacle for an unbroken information flow in the building process from the interviewed project managers point of view is the archiving, as it is not yet digital. It is also something that is investigated, and has been investigated for several years, but a good solution has not yet been found (Project manager C). One proposed solution to make the handover more efficient and digital is e.g. a platform where coordination can take place within the projects in order to avoid misunderstandings and make information available to all within the project. Where everyone can see what has been delivered and when (Project manager C).
CHAPTER 4. RESULTS 31
4.4 Summary
Figure 4.4: Summary of all results.
Chapter 5
Analysis & Discussion
The aim of this study was to investigate what obstacles that exist in the han- dover of geographic data from a building project to a management organiza- tion. The results have proven that there are factors in the handover process that affect the quality of geographic data, and the findings also highlights opportu- nities to make the handover more efficient. The study was delimited to focus on geographic data, and all the work and the analysis of the results have hence been made and interpreted in the context of geographic data.
In the analyzed data some themes including important aspects could be iden- tified. Further themes were also present, however most of those can be linked to or included in the main themes. The analysis derived from the results is also presented in relation to findings in previous research.
The main results are related to requirements and communication. A GIS database require a certain structure of files imported, regarding e.g. layers, lines and surfaces. For designers it is more important that the drawings look good on a paper, not the actual content such as line types and layer names. When the interest of the designer is not on the management phase and the ability to smoothly store the created objects in a GIS database, clear and early commu- nicated requirements are important in order to ensure that the designer knows what the management organization needs. With clear requirements, the issue has been solved in theory, but during the work the author of the study has re- alized more and more that the technology is not the most complicated factor, but the ones involved and communication between them. Through earlier in- volvement of the management organization communication can be improved, and knowledge and understanding of the relevance and importance of quality-
33
assured geographic data can be increased. These findings are further analyzed and discussed later in this chapter.
The study confirms previous research in the area of handovers, and it can be concluded that there are similar types of problems and obstacles and hence solutions that can be applied to streamline the handover in projects of vary- ing degrees. The findings and recommendations of this study could probably also be applied to other types of data, which means that this is a problem that should be of interest regardless of self-interest. As this study focus on geo- graphic data applications on other data will not be further discussed in this report. However, it can be seen as an important factor and an advantage also in the context of geographic data as it might mean more incentives and moti- vation from various sources to make the handover process more efficient.
All interviews gave valuable information and contribute to the findings of the study, with some degree of variation. If comparing what the different groups interviewed highlights, the designers do mostly focus on the requirements and the system managers on the communication. This does make sense. The de- signers are dependent on the requirements in order to being able to deliver what the management organization/client needs. The management organiza- tion, who probably think that the requirements are clear, believe that it is a question of communication and wish to be more included during projects and in handovers in order to make sure that there is more of a life cycle perspective and that the value of qualitative geographic data is understood. The project managers on the other hand did not really contribute to much of the findings regarding the requirements and preventing factors in general. This could still be contributing to the results and findings of this study, as the project managers seem to be part of the obstacles themselves. The project manager is expected to make sure that the project hands over what the client has ordered, but that seems to not always be the case.
CHAPTER 5. ANALYSIS & DISCUSSION 35
5.1 Communication
The management organization is often not involved in the early stages of a project, which means that they can have difficulties in setting requirements, checking, reviewing and approving digital deliveries. The interviewees see a strong value in involving the management organization early in the project to improve the communication and get each other’s perspective. This is also something that is mentioned in literature. Studies have been made on how early involvement of the management organization can affect the work and how it can be performed. Such a study was done by Whyte et al. [16] at the London Olympics, where early involvement of people with extensive experi- ence in management was provided, with good results as the interests of the management could be ensured.
System manager E and F describes larger projects where they have had the pos- sibility to meet the involved in the project at an early stage as successful from the management organizations point of view. Designers A and B mentions an- other big project where early workshops together with representatives from the management organization enabled to initiate a dialogue about the end result.
Early involvement of the management organization, and especially personal meetings, could prevent problems that arise as a result of the long information chains, and distance between e.g. designer and system manager, mentioned by Designer A since it can be made clear who is responsible for what. Early involvement of the management organization requires that the handover and management process is considered early on in the project, which committed project managers and committed operators mentioned by Mäkelä [20] could be a major contributing factor too.
The project is responsible to deliver data according to the client’s require- ments, but who is actually responsible to ensure that the data delivered has been checked against the requirements is unclear. The management organi- zation sets requirements and expects the data delivered to comply with those requirements (System manager E and F). The project manager expects that the requirements have been checked and will be checked but does not take respon- sibility for the control (Project manager C and D). To avoid this, everyone’s roles and responsibilities need to be made clear from the start.
5.2 Requirements
A prerequisite for effective handovers is, according to Almqvist et al. [4], that requirements do not only need to exist, they must also be followed. That clear requirements are communicated, and that they should be followed, can be seen as something obvious to make designers produce data that the system man- agers want to receive, but the designers interviewed testify that this is not always the case (Designer A and B). The fact that lack of requirements can be a result of lack of competence occurs both in previous research and in the results of this study. If the management organization or client does not have enough knowledge of the files or the format of the files that are requested, it can lead to requirements being imposed which the operator has no interest in or cannot meet. This can lead to frustration for both the management as well as the designers (Designer A). Winch [17] confirms that it can be difficult for clients to provide adequate requirements if they have less competence in the area than the contractor and Linderoth et al. [14] highlights this lack of compe- tence from the management organization, which leads to difficulties in making demands on contractors, as a main factor that constrain digital transformation.
The mutual frustration is confirmed by system manager E who on the other hand claims that in many cases, a holistic view is lacking where there is no understanding of the usability of the data after completion of the project.
Apart from the fact that the requirements are clear, another success factor men- tioned is that they are communicated early (Designer A). If the requirements are not clear or complete enough there is room for interpretation, which can lead to different results depending on the person interpreting. Schneider et al. [15] suggests that this should be prevented already in the contract, as more detailed specifications of work as part of the contract would lead to less risk for interpretation. Communicating the requirements early is also something that Whyte et al. [19] recommends to assure that the data provided has the desired quality and format. The fact that the requirements from the Traffic Adminis- tration Office are stated in a late phase of the technical manual assumes that the whole manual is read through at an early stage. This is something that the system managers at the Traffic Administration Office are critical of them- selves, as they see it is a critical step for the design of the geographic data but reading it all cannot be expected to be done (System manager E). When the designing is done, it is often considered too late to make changes, given the time frame and the budget of the project, which indicates that time and budget is considered as more important than quality-assurance of the data.
CHAPTER 5. ANALYSIS & DISCUSSION 37
5.3 Digitalization & Technology
The digitalization is a very topical subject right now, and there are several projects going on with the aim of making parts of the building process more digital and efficient. Both designers, project managers and system managers see advantages with more digitalized or automated processes.
The largest obstacle for a more digitalized handover process is mentioned by Project manager C and D as the archiving, this because the archive system cur- rently only can receive and handle paper copies. The archiving aspect is only mentioned as an obstacle by the project managers, which may be explained by the fact that those are the ones delivering the copies to the archive and hence taking it into consideration.
In the interviews a need for better control of the data in comparison to the requirements was identified. This because it is very time-consuming for the system managers to check the files handed over, and the checking is apparently not properly managed in the projects (System manager E). This, together with requests for a preventive step in the handover process, where project managers should not be allowed to deliver data that does not comply with the require- ments (System manager E), motivates potential improvement through techni- cal solutions.
5.4 Case - Traffic Administration Office
A more digital process and increased life cycle perspective do not need to mean that all processes need to be completely automatized or digitalized. In the case of the Traffic Administration Office’s management of park data, a clear oppor- tunity and need can be identified from the result of this study. That the system managers manually edit and enter the geographic data into the database is both time consuming and unnecessary duplication work. Such time-consuming tasks could be eliminated with a process that automatically checks the data against the requirements when delivered, and refuses reception unless the data complies with the requirements. In order to enable the requirements to be automatically and digitally comparable to a delivered file, the requirements,
which in the case of the Traffic Administration Office are in text format, must also be converted to parameters.
Technical solutions such as an automatic control still require that the require- ments are communicated and considered. As previously mentioned in the re- sults and the analysis, it is of great importance that the requirement setting is specific, so that the system managers gets what they have demanded and needs. To do this it is important that those making the requirements have the right competence for the requirements to maintain an appropriate level. To assure this, it could be beneficial to involve experienced designers in the re- quirement setting. Regardless, the requirement setting need to be evaluated if it seems to not be adequate, in order to be improved. That system managers are involved earlier in projects and the handover process also seems to be a win- ning concept. Closer cooperation and better communication between, above all, the Stockholm City Development Administration and the Traffic Admin- istration Office which are involved in ordering, execution and management of construction projects may be needed to enable this. Improved internal orga- nization is also requested and highlighted by the project managers and system managers interviewed.
In addition to the fact that the designers should follow the requirements given and the project managers should ensure that the requirements are followed dur- ing the work and in the handover, the system managers must also act. In the result it appears that those responsible for the management themselves is con- tributing to the quality loss, as they manually enter the data into the database instead of giving feedback to the one who delivered the data.
One factor only mentioned briefly, and which is probably most difficult to in- fluence and prevent, is the human factor. Although there are requirements and tools that are intended to facilitate and make things more efficient, they are threatened by the fact that one can forget, does not consider it important, can- not use it fully, or similar causes. It is hard to find a solution to this, but it could possibly be prevented by clear routines, which also requires closer collabora- tion and better contact between the different units, and better understanding of each other’s needs. As Almqvist et al. [4], among others, highlights a holis- tic view is an important factor, which can be applied in large as well as small projects and organizations, and is something that needs to be worked on both internal at the Traffic Administration Office as well as external in projects in general.
CHAPTER 5. ANALYSIS & DISCUSSION 39
5.5 The handover process
For a sustainable society and smart cities, the building industry must develop and become more efficient with uninterrupted information flows. Although handing over and managing geographic data is an important part of an un- broken information flow, few studies are available and little attention is given.
The reason for this cannot be ensured, but one reason to that could be that it is not seen as a process itself, but more part of other processes. Also, cur- rent projects with focus on streamlining the building process focus mainly on earlier stages in the process, which will possibly change with future increased digital maturity in the whole information chain.
5.6 Data collection
An interesting aspect of the results of the qualitative interviews is, as previ- ously mentioned, that the project managers did not really contribute to much of the findings regarding the requirements and preventing factors in general.
Another important aspect, however, is that this result may be directly depen- dent on the individual project managers who were interviewed. It cannot be denied that other project managers could have contributed more or given other answers. However, this implies to all interviewees and is always a risk in qual- itative interviews.
As mentioned in Section 3.3.2 Interviewee Selection only six persons were interviewed. This can be considered as a too small selection to be able to give a representative and generalized result and hence a disadvantage for the research. The fact that the selection of interviewees was not larger is due to late response from the respondents contacted, or that they were unable to book an interview until much later, which were prevented by the time frame of this thesis. This could possibly have been avoided by gathering interviewees at an earlier stage, but on the other hand it is not certain that all respondents would have been available earlier anyways, and the outcome would have been the same. Another aspect that limited the selection of interview subjects and made the process of obtaining suitable candidates difficult is the connection to the Traffic Administration Office in Stockholm and park data. In order to de-
limit the study, the Traffic Administration Office in Stockholm’s reception of park data was chosen as a case study, and only designers and project managers who has supplied data to the database with geographic data representing park objects were contacted.
41
