Collaboration, experimentation, continuous improvement? : a critical exploration of an agile way of working in the planning department context

(1)

COLLABORATION,

EXPERIMENTATION,

CONTINUOUS

IMPROVEMENT?

A critical exploration of an agile way of working

in the planning department context

Trey Hahn

(2)

A critical exploration of an agile way of working in the planning

department context

Master Thesis

University of Amsterdam

Graduate School of Social Sciences

Urban and Regional Planning

Dilemmas of urban mobility, and beyond

Supervisor:

Marco te Brömmelstroet

Second Reader:

Tuna Tasan-Kok

Word Count: 15,150

(main body of paper without figures/tables or footnotes)

Trey (Harold) Hahn: #11191457

11 June 2019

(3)

Abstract

Urban planners aim to create better living environments for people, and planning

departments move towards this goal on a project level. Thus, project execution is key for planning to achieve its purpose. This multi-method research examines the way of working within planning departments and explores how they might transition to working on projects in a way that better serves citizens while using less money and time. It uses the Programma Fiets (Bicycle Program) of the Gemeente Amsterdam (Municipality of Amsterdam) as a case study, and “agile” to operationalize the way of working. It utilizes semi-structured interviews with practitioners along with a literature review, specific analysis of the Alexanderplein intervention, and guided questionnaire sessions as its methods. The research begins by building a holistic understanding of agile in the context of the planning department. It then looks at the gaps between the current way of working in the Municipality of Amsterdam and agile before testing potential solutions to overcome barriers to an agile way of working in practice. At the end, the researcher reflects on limitations and proposes future research.

(4)

Thesis structure

The thesis is structured as follows:

Chapters 1 and 2 include context for the research and its theoretical base. First, an

introduction on (problematic) project delivery is given and the question what is a successful

project is posed. Then, the theoretical framework of a multi-level perspective on transitions is explained, a research question and sub-questions are derived around it, and an agile way of working is brought in as a way to operationalize the research.

Chapters 3, 4, and 5 report on the research itself. Chapter 3 explains the case and the methods chosen to answer the research (sub-)question(s). Chapter 4 presents findings from each method to answer the sub-questions. Chapter 5 brings it all together in a conclusion.

Chapter 6 reflects on the research. As a student researcher doing a master thesis, reflection is critical- this section notes limitations and looks into avenues for future research. Finally, Chapter 7 lists references and Chapter 8 contains the appendices.

Acknowledgements

This research would not have been possible without the support, guidance and time of many people. First, thank you to all of the practitioners that took the time out of their days for the interviews and survey sessions. It has been a privilege to hear you reflect on your work- it makes me feel inspired and hopeful! Thank you to my thesis supervisor, Marco te

Brömmelstroet- you’ve helped me ask questions and be curious: a great environment in which to learn and do a thesis. Thank you to my family for supporting me from abroad, and to my friends here in Amsterdam- it has been really great to have people to talk to and lean on for support during the research process, emotional just as much as academic.

(5)

CHAPTER 1

(7)

1. Introduction

“NYC's brand new subway is the most expensive in the world — that's a problem” read the headline of one major media outlet after the long-awaited opening of the first phase of the Second Avenue Subway in New York City (Yglesias, 2017)⁠. Others were not much nicer: one

said “It Took a Very, Very Long Time for the Second Avenue Subway to Be a Reality” (Cullen,

2017) and another “The Insanely Expensive Second Avenue Subway Explained” (Pope-Sussman, 2016).

New York City waited almost 100 years for just the first phase of the project, and paid billions 1

of dollars more than it bargained for: “To put the $4.5 billion cost of Phase 1 in context, when

the Second Avenue Subway was first proposed in the late 1920s [originally 3 phases], the

estimated price tag of the entire project... works out to about $2.4 billion in 2016 dollars.

That's approximately what one mile of Phase 1 cost” (Pope-Sussman, 2016, paragraph 10).

Unfortunately, this is not an isolated incident. Phase two of the project is expected to cost even more, with a $6 billion price tag translating into about $2.2 billion/kilometer (Yglesias, 2017), and the estimated costs of another rail project in New York (“East Side Access”) have grown to “$12 billion, or nearly $3.5 billion for each new mile of track” (Rosenthal, 2017, paragraph 5). A project on the neighborhood level in the west side of Manhattan in New York has a local community board member frustrated and confused: “‘They’re going to spend $200 million, and we’re not getting anything,’” (Meyer, 2018, paragraph 3)- despite connecting to the busiest bike path in the country, the project doesn’t include basic protection for people cycling (Meyer, 2018). In Amsterdam, the Noord-Zuidlijn project finally was completed in 2018

at a cost of “more than three times as much as estimated” (Van Leeuwen, 2018, paragraph 4).

It was originally scheduled to open in 2005 (“Amsterdam North-South Metro Line Opens 22 July 2018,” 2018).

In the academic realm, there is extensive literature on cost overruns in planning projects (Cantarelli, Flyvbjerg, Molin, & Wee, 2010; Flyvbjerg, 2007; Flyvbjerg, Holm, & Buhl, 2003, 2002, 2004), and while the Second Avenue Subway is an extreme case (Nasri, 2013; Plotch, 2015)⁠, challenges in execution of projects are happening everywhere. This raises larger

universal questions: What is a successful urban planning project? How can planning

institutions work in order to make sure their projects are successful?

1.1 What is a successful project?

In order to investigate these questions, this paper draws on both the fields of project management and urban planning. There is already much literature on the definition of a successful project around both planning infrastructure projects (Flyvbjerg, 2014; Giezen, 2012; Gutenschwager, 1973; Lehtonen, 2014) and project management at large (Alias, Zawawi, Yusof, & Aris, 2014; Attarzadeh & Ow, 2008; Davis, 2014; DeWit, 1988; Hussein, Ahmad, & Zidane, 2015; Müller & Jugdev, 2012; Radujković & Sjekavica, 2017; Shenhar, Levy, & Dvir, 1997). However, despite extensive research there is no consensus. Defining project success is

1_{To give perspective on how much time the project took, one writer noted that “the Chicago Cubs have}

won a World Series and a man has walked on the moon. Fifteen governors and another 15 mayors have come and gone over that time” (Cullen, 2017, paragraph 3)

(8)

“a subject of controversy” (Karami & Olatunji, 2018, p. 122) and “the only agreement seems to be the disagreement on what constitutes [it]” (Prabhakar, 2008, p.3).

While it’s clear that success can mean different things to different people (Jugdev & Müller, 2005), there is a well-used theoretical base in the project management field: the iron triangle (Figure 1). The iron triangle measures success through performance according to schedule, cost, and quality of a project (Jha & Iyer, 2007). There has been significant criticism of it (Dimitriou, Ward, & Wright, 2013; Toor & Ogunlana, 2010) for not fully encompassing all of the nuances of project success. This research acknowledges that it is an imperfect theoretical base and that “project success is dependent on one’s perception and perspective” (Ika, 2009, p. 7).

Figure 1: The “iron triangle” of project management (Source: Ebbesen & Hope, 2013, p. 2)

However, the iron triangle has still “become the de-facto method to define and measure project success” (Ebbesen & Hope, 2013, p. 2). It is well-known (Atkinson, 1999; Ika, 2009) and there is power in its simplicity. As this research is not suited to provide a definitive answer to the larger debate on the definition of project success, it will instead adapt the iron triangle to the context of the research.

Of its three corners, cost and time are apparent in planning projects and straightforward to measure. The “quality” of a project, however, is not easy to measure in urban planning. There needs to be an indicator more suitable to the planning context, and one that widens the scope of success to be more encompassing (a weakness of the iron triangle). To achieve this, the betterment of life for citizens is a suitable choice. It is a central concept of relevance for

urban planning and the reason why the profession came to be in the first place (Hall, 1996). 2

This leads to the adapted indicators of success for the purpose of this research: time, money, and better serving citizens. “Better serving citizens” is hard to measure and certainly can vary depending on the context, so this paper takes a descriptive, qualitative approach to it based

on what was learned during the research.

2_{As one professional stated reflecting on 60 years of experience in the profession, “the ultimate role of}

the planner is to help a community become a better place” (Bolan, 2016, p. 285). Other scholars have noted that planning is “an idea made up of concepts and sets of practices, which aspire to change the world for the better” (Campbell, Tait, & Watkins, 2014, p. 47).

(9)

CHAPTER 2 THEORY, RESEARCH

QUESTIONS, AND

OPERATIONALIZATION

(10)

2. Theory, research questions and operationalization

In order for a planning department to complete projects with less time and money spent while better serving citizens, the current way of working has to change. Zooming out, there is a way of understanding transitions on a higher level through a multi-level perspective (MLP) as conceptualized by Geels (2002). There has been much academic use and discussion of this framework (Jørgensen, 2012; Moradi & Vagnoni, 2018; Papachristos, Sofianos, & Adamides, 2013; Smith, Voß, & Grin, 2010; Whitmarsh, 2012)⁠, and extensive elaboration by its author (F.W. Geels, McMeekin, & Pfluger, 2018; Frank W. Geels, 2011, 2012, 2018; Frank W. Geels,

Schwanen, Sorrell, Jenkins, & Sovacool, 2018). While it was originally used to understand technological transitions and outcomes, in this research it is applied to the context of transitioning a way of working.

There are 3 levels in the MLP: a landscape, regimes, and niches (Figure 2). The landscape comprises the external environmental factors of what is happening overall, and it influences everything else. The regimes operate beneath it. They both influence and are shaped by development of the landscape, and also strongly influence the level below them, niches. Niches are the bottom level of the MLP where new ideas start. Those ideas can come up and potentially influence both the regimes and landscape, while at the same time their own development is highly influenced by what is going on at the top two levels.

Figure 2: Hierarchy of a multi-level perspective on transitions (Source: Geels, 2002, p. 1261)

Many of these new ideas (“novelties” as described by Geels) die out, while others evolve and are incorporated into a regime, and some eventually go on to alter the landscape (Figure 3). It should be noted that change is not easy, and many good ideas will fail to reach the regime level.

(11)

Figure 3: Potential paths for novelties (Source: Geels, 2002, p. 1262, via Rip & Kemp, 1996, and Kemp, Rip, & Schot, 2001)

Relating the MLP back to context of the way of working in a planning department, the existing way of working can be thought of as a regime, and the idea for a different way of working lies below in a niche (Figure 4). A goal of this research was to investigate how this different way of working can come up and be adopted by a regime (i.e. planning department) or eventually even alter the larger landscape.

Figure 4: Planning department and way of working in an MLP (Adapted from Geels, 2002, p. 1261)

2.1 Research questions and sub-questions

From the literature review on what a successful urban planning project entails and the MLP on transitions (logic in Figure 5), an overarching question was developed to guide the research:

How can planning departments transition to working on projects in a way that uses less money and time while better serving citizens?

(12)

Figure 5: Logic leading to research question

The next step was to operationalize spending less time and money while better serving citizens into a way of working. Due to its extensive literature (Dingsøyr, Nerur, Balijepally, &

Moe, 2012) and widespread use in IT and project management fields, “agile” was chosen for 3

this. It is a collaborative, iterative way of working that has become a very popular way to execute projects and provide more value for the customer (or in the planning department’s case, the citizen) using less time and money (Dybå & Dingsøyr, 2009; Highsmith, 2002; Pinto & Serrador, 2015). It has become a niche way of working in urban planning and elements of it are present in projects in Seattle (Schmitt, 2019), Singapore (Scruggs, 2018) and Amsterdam (Wagenbuur, 2018). Agile aligns with the indicators of time, cost, and citizens (Cockburn & Highsmith, 2001) in the research question, and is explored more in the following sub-section.

Conceptually, an agile way of working has to rise up into the planning department’s regime. While parts of it have been utilized in particular contexts, agile has not been adopted on a widespread level. In order for that to happen, it has to be accessible and relatable to the regime (Figure 6). In a competitive environment where many novelties die out, an idea is not likely to be adopted by a regime without fitting into its context.

Figure 6: Agile and planning department in an MLP (Adapted from Geels, 2002, p. 1261)

3_{This research chooses to write agile in sentence case to avoid the over-commercialization and hype}

that has developed with the term (see Morris (2017) for an elaboration on this thought). Quotations from external sources maintain the original author’s text.

(13)

This graphic representation led to logical theoretical steps to answer the research question. The first step was to enable agile to rise into the regime by making it understandable to the planning department. The next step was to better understand the context of the planning regime, before lastly searching for solutions that enable an agile way of working to rise up into it (or even alter the larger landscape). These steps were translated into sub-research questions to structure the research project, and can be found below in Table 1.

Theoretical step Sub-question

1 Make agile detectable by wider planning regime

How accessible and relatable is agile to planning

departments?

2 Better understand context of planning department

What are gaps between the current way of working and agile in planning

departments?

3 Test potential solutions to have agile rise up into the planning regime

Under what conditions might agile be adopted by planning departments in practice?

Table 1: Theoretical steps & sub-questions (images adapted from Geels, 2002, p. 1261)

2.2 Agile and urban planning

Agile

Agile is “the ability to create and respond to change... a way of dealing with, and ultimately succeeding in, an uncertain and turbulent environment” (“What is Agile Software

Development?,” n.d., paragraph 1). It focuses on delivering a working product, collaborating with customers, valuing individuals and interactions, and responding to change (Beck et al., 2001). It is a more iterative and incremental way of working on projects in comparison to the

(14)

traditional “waterfall” approach (Shawky, 2014). Table 2 describes an agile perspective in detail, and compares it to a traditional view.

Table 2: Difference between traditional and agile way of working (Source: Dybå & Dingsøyr, 2009, p. 7, via Nerur & Balijepally, 2007)

The planning context

Given that the goal of planning is to better serve citizens (Bolan, 2016), the potential of agile to facilitate projects that are more aligned with people’s needs using less money and time (Cockburn & Highsmith, 2001) is directly relevant to planning. Despite that relevance, agile has not yet seen widespread adoption in planning. If one conceptualizes an agile way of working as a tool, the case of planning support systems (PSS) (Geertman & Stillwell, 2004; Geertman, Toppen, & Stillwell, 2013; Klosterman & Pettit, 2005)- which are essentially tools to help planners do their complex job better (te Brömmelstroet, 2013)- can provide insight into its adoption. The significant challenges in adoption of PSS by planners (Vonk, Geertman, & Schot, 2005) have led scholars to reflect on why they haven’t been adopted and how they can so that the end goal of improving planning practice can be reached (te Brömmelstroet, 2017).

Agile in planning?

While it has roots in multiple contexts (Abbas, Gravell, & Wills, 2008; Freedman, 2009; Highsmith, 2002; Rigby, Sutherland, & Takeuchi, 2016a), agile may be best known for its rise

in the field of software development. Some authors are very excited about it and see 4

potential for it to spread to other industries: “Agile innovation has revolutionized the software industry, which has arguably undergone more rapid and profound change than any other area

4_{In 2011, less than 10% of major US federal government IT projects were self-described as “Agile” or}

“iterative” (Viechnicki & Kelkar, 2017). Yet in 2017 that number rose to an astonishing 80% (Viechnicki & Kelkar, 2017). According to the Project Management Institute, “organizations increasingly embrace agile as a technique for managing projects” (Pulse of the Profession: 9th Global Project Management Survey, 2017).

(15)

of business over the past 30 years. Now it is poised to transform nearly every other function in every industry” (Rigby, Sutherland, & Takeuchi, 2016b, p. 50)⁠. Others are less bullish, arguing that “agile methods are neither panacea nor silver bullet” yet note that more research is needed due to the small sample size of their study (Budzier & Flyvbjerg, 2013, p. 22).

Organizations have already started to use an agile approach in multiple fields

(Narayanamurthi, 2017), and academic explorations have occurred on the application of agile principles to fields such as education (Andersson & Bendix, 2006; Lang, 2017), construction (Nowotarski & Pasławski, 2016; Streule, Miserini, Bartlomé, Klippel, & García de Soto, 2016), health care (Tolf, Nystrom, Tishelman, Brommels, & Hansson, 2015), and marketing (Poolton, Ismail, Reid, & Arokiam, 2006).

In urban planning practice, elements of an agile way of working are already emerging (Schmitt, 2019; Scruggs, 2018; Wagenbuur, 2018). There have also been several conceptual explorations of agile related to planning. Munro (2015)⁠ sees agile through the lens of

technology and smart cities, while Jin & Stough (1996) see intelligent transportation systems and technological infrastructure as one part of the larger idea of an agile city. W. W. Clark (2007)⁠ describes agile energy systems and sustainable communities, and Luna, Kruchten, & Moura (2015)⁠ construct a theory of agile governance. Velibeyoglu, Sargin, Bingöl, Saygin, & Yildiz (2016) explore the application of an agile framework to urban design. Another group of authors directly translates the principles of the agile manifesto (Beck et al., 2001) to urban adaptation (Pathirana, Radhakrishnan, Ashley, Quan, & Zevenbergen, 2018; Radhakrishnan, Pathak, Irvine, & Pathirana, 2017).

These are all valuable conceptual contributions around around applying agile to urban

planning. However, what remains unanswered by them is how agile translates on an everyday practical level to the context of the planning department. That’s where this research comes in.

2.3 Operationalizing agile

There are many different perspectives on agile, and for this research some decisions had to be made up front about how to define it. Lists summarizing the parts of agile relevant to the research (sub-)question(s) were made, which were then used throughout the research.

In order to create the lists, literature on what an agile organization is and does, along with barriers to its implementation, was consulted, and a smaller group of main sources were selected. These sources were read in more detail, and key elements on the aforementioned subjects were noted down. The key elements were brought together and consolidated to create three lists: characteristics of an agile organization, practices of an agile organization, and barriers to the adoption of agile. The first two lists relate primarily to sub-question 2 (exploring how the existing planning department works, and then comparing it to agile), while the last one relates to sub-question 3 (understanding conditions- and the barriers needed to be overcome- for agile to be adopted in practice). They lists were used in the semi-structured interviews in the second part of the sessions to get the interviewee thinking about how they relate to their working context, and in the development of the questionnaire.

(16)

It should be noted that the initial list was slightly revised after feedback from the first

interviewee about some parts being less clear than others (which made sense- the lists were developed largely from IT and project management literature, so some of the language used may have not been familiar in the planning context). After minor revisions, the lists were made final and used throughout the rest of the research. The final lists on characteristics (Table 3), practices (Table 4), and barriers (Table 5) are below.

Characteristic Sources

Focus on people and interactions over processes and tools

Beck et al. (2001), Cockburn & Highsmith (2001), Highsmith (2002)

Responsive Cockburn & Highsmith (2001), Dybå & Dingsøyr (2009), Nerur & Balijepally (2007)

Strong focus on satisfying customer Beck et al. (2001), Nerur, Mahapatra, & Mangalaraj (2005)

Can respond to (and welcomes) change Beck et al. (2001), Nerur & Balijepally (2007) Self-organizing teams Beck et al. (2001), Cockburn & Highsmith (2001),

Highsmith (2002), Nerur, Mahapatra, & Mangalaraj (2005)

Flexible, organic organizational form with interchangeable roles

Nerur & Balijepally (2007), Nerur, Mahapatra, & Mangalaraj (2005)

Foster individuals’ skills and encourage creativity while working together as a team

Beck et al. (2001), Cockburn & Highsmith (2001), Highsmith (2002), Nerur, Mahapatra, & Mangalaraj (2005), Nerur & Balijepally (2007)

Information drives decisions, not a work hierarchy

Cockburn & Highsmith (2001), Nerur & Balijepally (2007)

High level of individual autonomy Dybå & Dingsøyr (2009), Nerur & Balijepally (2007) Technical excellence Beck et al. (2001), Highsmith (2002)

Embraces conflict and discussion, blends chaos and order

Highsmith (2002), Nerur & Balijepally (2007) Table 3: Characteristics of an agile organization

Practice Sources

Frequent collaboration between roles within project team and with customer

Beck et al. (2001), Cockburn & Highsmith (2001), Highsmith (2002), Nerur, Mahapatra, & Mangalaraj (2005)

Reflection and adjustment at regular intervals

Beck et al. (2001), Highsmith (2002), Nerur & Balijepally (2007), Nerur, Mahapatra, & Mangalaraj (2005)

(17)

Early, incremental and continuous delivery of a working product

Beck et al. (2001), Cockburn & Highsmith (2001), Highsmith (2002)

Continuous testing & improvement Highsmith (2002), Nerur, Mahapatra, & Mangalaraj (2005), Nerur & Balijepally (2007)

Frequent feedback Nerur, Mahapatra, & Mangalaraj (2005), Highsmith (2002)

Simplification- maximize the amount of work not done (only spend time on things that deliver value to customer)

Beck et al. (2001), Highsmith (2002)

Experimentation, iterative work Nerur & Balijepally (2007) Manager is facilitator Nerur & Balijepally (2007) Maintain a constant work pace Beck et al. (2001)

Table 4: Practices of an agile organization

Barrier Sources

Lack of effective collaboration and teamwork

Chan & Thong (2009), Gandomani & Nafchi (2016), Nerur, Mahapatra, & Mangalaraj (2005)

Organizational culture (resistant to change) Gandomani & Nafchi (2016), Nerur, Mahapatra, & Mangalaraj (2005)

Management style Nerur, Mahapatra, & Mangalaraj (2005)

(New) skills, abilities, knowledge needed Chan & Thong (2009), Gandomani & Nafchi (2016), Nerur, Mahapatra, & Mangalaraj (2005)

Shared understanding of agile (mindset) Chan & Thong (2009), Gandomani & Nafchi (2016) (Detrimental) perceptions of agile

methodology

Chan & Thong (2009), Gandomani & Nafchi (2016), Nerur, Mahapatra, & Mangalaraj (2005)

Reward systems & employee career consequences

Chan & Thong (2009), Nerur, Mahapatra, & Mangalaraj (2005)

Organizational structure Nerur, Mahapatra, & Mangalaraj (2005) Customer relationships and external

conditions

Chan & Thong (2009), Nerur, Mahapatra, & Mangalaraj (2005)

Existing technologies/infrastructure Nerur, Mahapatra, & Mangalaraj (2005) Top management support Chan & Thong (2009)

(18)

CHAPTER 3

(19)

3. Research design

An explorative research with multiple methods was conducted to answer the three

sub-questions, which together attempt to answer the overarching research question of how planning departments can work in a way that better serves citizens while reducing the time and money spent on projects.

3.1 Case study

In selecting a case study, there are multiple factors to consider; for example: relevance to the research question, research feasibility (practicalities), and relevance for society. Additionally, the research object must be a reasonable size to ensure that the research is feasible. With all this in mind, the Bicycle Program of the Municipality of Amsterdam (“Programma Fiets” in the “Gemeente Amsterdam”, in Dutch) was selected as the planning department case to study the

research questions . The Municipality of Amsterdam is arguably on the cutting edge of 5

planning, and has been described as a “pioneer in technological and social innovations” (Vasarini Lopes, 2018, p. 20). Recent projects such as the redesign of Mr. Visserplein

(Wagenbuur, 2018) and the “Ping If you care!” initiative (Mobiel21 & Bike Citizens, 2019) show the city is looking to make infrastructure that works for citizens and is open to new ways of approaching projects (Copenhagenize Design Co., 2014; Gemeente Amsterdam, 2018).

Within the municipality, the bicycle program is a crucial part of the city’s larger mobility plans (Gemeente Amsterdam, 2013). It also has larger implications for learning about planning for increased cycling, which is relevant because of the plethora of societal benefits of cycling that recent academic literature has illustrated (Fishman, Schepers, & Kamphuis, 2015; Garrard, Rissel, & Bauman, 2012; Gössling & Choi, 2015; Gotschi, 2011; te Brömmelstroet, Nikolaeva, Glaser, Nicolaisen, & Chan, 2017; Woodcock et al., 2009). Practical access to the research object was also logistically convenient as the researcher resided in Amsterdam.

While there are many different levels and types of planning departments in the world, in order to make the scope for this research manageable the aforementioned case has been chosen to study in more detail. What is learned can be used to contemplate larger implications.

(20)

3.2 Methods

The methods that were used to study the research questions are outlined in Table 6, and below they will be explained in more detail. Figure 7 illustrates the logic followed to arrive at the sub-questions from the main research question, and the methods selected for each sub-question.

Sub-question Data Collection Method

Data Analysis Method

1. How accessible and relatable is agile to planning departments?

Literature review Academic papers Literature synthesis matrix 2. What are gaps between

the current way of working and agile in planning departments? Narrative interview. process mapping session, online content search Narrative interview transcripts, process maps, results from online content search

Thematic analysis on transcripts; together with all 3 sub-methods: “assisted process analysis”

3. Under what conditions might agile be adopted by planning departments in practice? Guided questionnaire sessions Questionnaire responses, researcher notes from sessions

Basic analysis from questionnaire responses and sessions All 3 sub-questions* Semi-structured

interviews

Interview transcripts Thematic analysis of transcripts

Table 6: Research sub-questions and corresponding methods (*the semi-structured interviews address a targeted part of all 3 sub-questions)

Figure 7: Research logic and methods for each sub-question

(21)

3.2.1 Literature review (sub-question 1)

The first component of the research was a literature review (Torraco, 2005). Sub-question 1

(How accessible and relatable is agile to planning departments?) was broken into two more

targeted questions (Figure 8). The literature review targeted the academic perspective, while the semi-structured interviews targeted the practitioners’ perspective.

Figure 8: Sub-question 1 split into two targeted questions

To conduct the literature review on the academic perspective, a literature synthesis matrix (Clark & Buckley, 2017; Webster & Watson, 2002) was used to organize information. The idea of synthesis is distinct from analysis and is an important part of this method. Using a puzzle as a metaphor, Clark & Buckley (2017) explain the difference, “Analysis is taking an already completed puzzle apart, and synthesis is putting individual puzzle pieces together to

complete the puzzle” (p. 354). The goal of this part of the literature review was to take ideas from different disciplines and put them together to reach new insights. They are contexts that may appear disparate at first glance, yet upon further examination there are key connections. That is what this review did- it aimed to look beyond the surface conclusions of the authors and conceptualize deeper insights.

Selection of literature

To start, a very wide informal search was conducted around agile and urban planning. After this wide search, three categories were identified that targeted the sub-question through multiple perspectives and related back to the main research question. They are as follows:

● Planning support systems (PSS): what planning departments need for a tool to be accessible/relatable and successfully integrate into their workflow (note: an agile way of working can be thought of as a PSS that needs to be integrated into practice) ● Agile organizations: characteristics and practices that are present in agile

organizations; barriers to adoption of agile

● Translating agile: relatability of agile to urban planning context; how other fields have taken up agile

Two pieces of literature from each category (six articles in total) were selected for the synthesis matrix. The individual rationale for each piece is given in Table 7 below.

(22)

Category Source Rationale 6

1. Planning support systems (PSS)

Improving the adoption and use of planning support systems in practice - Vonk G, Geertman S (2008)

Insights from the PSS implementation gap can help inform about the context of the planning department. The authors’ bottleneck categories of diffusion to and in planning organizations and user acceptance are especially relevant lenses. Transparency, flexibility,

simplicity: From buzzwords to strategies for real PSS improvement -

te Brömmelstroet M (2012)

This piece is grounded in the PSS implementation gap and gives 3 real-world planning department test cases. Practical insights on transitioning to an agile way of working can be gained from this. 2. Agile

organizations

Theoretical Reflections on Agile Development Methodologies -

Nerur S, Balijepally V (2007)

This piece elaborates on what an agile approach is and explains how it is developing in disciplines beyond software development.

Acceptance of agile

methodologies: A critical review and conceptual framework -

Chan F, Thong J (2009)

A survey of literature on the acceptance of agile methodologies and resulting conceptual framework based on a knowledge management perspective inform on barriers to implementing agile in practice.

3. Translating agile

Managing urban water systems with significant adaptation deficits—unified framework for secondary cities: part II—the practice - Pathirana A, Radhakrishnan M, Ashley R, Quan N, Zevenbergen C (2018)

A translation of the principles of the agile manifesto (Beck et al., 2001) to urban adaptation tells us about how relatable agile is to the context of urban planning. The case of the city of Can Tho relates this back to serving citizens.

Implementation of Scrum in the Construction Industry Streule T, Miserini N,

Bartlomé O, Klippel M, García de Soto B (2016)

This paper conducts a case study on the use of scrum , an agile methodology, in a construction 7

company. This offers a lens to reflect on how how accessible agile is to another profession and how it is actually adopted.

Table 7: Literature selected for literature synthesis matrix with rationale

Synthesis

The author used three layers of a literature synthesis matrix for his own purposes of

understanding and organizing the selected literature. The first layer was on conclusions of the articles related to two relevant themes, the second on who was brought to the table in the article, and the last on what makes the article unique. The two themes were taken from the

6_{Relevance to sub-question 1 and/or overarching research question}

7_{While scrum is a commercial methodology and is often capitalized, this research chooses to write it in}

sentence case to avoid mystifying it. Quotations from external sources still maintain the original text.

(23)

sub-question by breaking it into two broad parts: agile and the context of the planning

department. After organizing thoughts on the articles around these themes in the three layers, a final matrix was made to summarize what was learned (see findings section). In line with Van Wee & Banister (2016), added value of this review is in real-world applications (of how to approach planning organizations with agile) and empirical insights (on the nuanced nature of how accessible and relatable the concept is).

3.2.2 Semi-structured interviews (sub-questions 1, 2 & 3)

The next component of the research was interviews (Bryman, 2012, chapter 20; Weiss, 1995) of practitioners and thematic analysis (Braun & Clarke, 2006) of the transcripts. According to Guest, Bunce, & Johnson (2006), 6-12 interviews is an ideal saturation point. In this method, 12 semi-structured interviews were conducted with different stakeholders relevant to the way of

working in the municipality . As this was an exploratory research, semi-structured interviews 8

were an ideal method for “examining uncharted territory” with open-ended questions to get practitioners’ independent thoughts (Adams, 2015, p. 493-494). This method addresses parts of all three sub-questions, and is combined with another method for each sub-question (like described in the literature review section for the academic vs. practitioner perspective). The questions addressed by the interviews are as follows:

SubQ1 (targeted): How accessible and relatable is agile to planning departments from the practitioners' perspective?

SubQ2: What are gaps between the current way of working and agile in planning departments?

SubQ3 (targeted): Which barriers are most relevant to the adoption of agile by planning departments in practice?

For sub-questions 1 and 3, the interviews target part of the sub-question, while sub-question 2 is addressed in general. Differences in how each method targets their part of the

sub-question are explained their respective sections. This section is written as follows: rationale for selection of participants is given, and then interview design and the approach to analysis are explained.

Participant selection

In order to gain a rounded understanding of the current way of working in the Bicycle

Program of the Municipality of Amsterdam and the conditions that would allow agile to be 9

adopted within it, multiple stakeholders that affect the way of working in their own ways were interviewed. Rationale (how the role influences the way of working and/or relates back to the

indicators of the main research question) is given in Table 8 below. 10

8_{Of the 12 interviews, 11 were with people in the Municipality of Amsterdam and 1 was with someone}

from the Vervoerregio Amsterdam, a regional planning organization that works with the municipality.

9_{9 of the 12 interviewees were or had at one point been involved with the Bicycle Program of the}

Municipality of Amsterdam, as had the 1 interviewee from the regional planning organization. The remaining 2 worked in other parts of the municipality.

(24)

Role # Interviewees Rationale

Planner 3 Creates plans and influences what happens in projects Program manager 3 Works at a higher program level, sets example for

team of project managers

Project manager 2 Manages the project level of working, which has implications for time and money spent, and for overall project direction

Urban/traffic designer 2 Shapes physical design at smaller scale that affects citizens and the time and money spent

Community engagement 1 Has direct contact with citizens, can get their feedback Policy 1 Guides project content at a high level

Total 12

Table 8: Interviewee roles and rationale for selection

This selection was done with the idea in mind that there are many people that influence how a project is delivered to citizens, not just planners. It should be noted that while participants were categorized into one role for purpose of this research, in practice they sometimes overlapped into multiple roles, so there are limitations to these labels.

Interview design

The interview sessions began with an introductory section according to best practices practices such as explaining the purpose and giving an overview of the interview, asking for permission to record the audio, and assuring confidentiality and the ability to seek clarification of questions or to decline to answer a question, as identified in Whiting (2008, p. 37). There was a focus on making the interviewee comfortable and communicating that the interview was more of a conversation (and partnership as Weiss (1995) describes) than an interrogation. As stated by Leech (2002, p. 665), “it is possible to be honest without being scary”.

The first question asked was intended to “warm up” the interviewees (Zorn, n.d.): “What is your role and responsibilities here?”. From there, an item list (Table 9) guided the rest of the interview. The item list has questions that come from specific items, which fit into larger themes (specifically targeted sub-questions for each section indicated in table).

Theme Items Questions

Current way of working (SubQ2)

Organizational

characteristics About the group you work in at the gemeente: ● What is it called & approx. how many people work in it? ● What different roles exist in it?

● What projects does it work on? ● What is its goal?

Details on existing regime &

perception of

● What stands out to you the way of working in the group?

(25)

project success *To (1) you & (2) your group at gemeente ● money spent on a project? ● time spent on a project? ● serving citizens in a project?

Project level Can you tell me the story of a recent project you’ve worked on that you thought had a effective (or not) way of working?

● How was the process from start to finish? ● Who were the people involved?

● What was your role?

● What was the project’s goal? ● What was the end output? Reflection on project:

● What do you think was most successful about it? ● What do you think could be improved?

○ How would you suggest improving it? Agile intro

questions (SubQ1)

Exploring one way of working

● Have you heard of the term “agile” before?

● What does an “agile” way of working mean to you? Agile

(SubQ2&3) Characteristics (SubQ2) Interviewees identify from lists (1) what is relevant for planning, (2) what reminds them of a project in their workplace Practices (SQ2) Storytelling goes from there on each item they bring up Barriers (SQ3)

Table 9: Semi-structured interviews item list

After exploring the interviewee’s organization, perceptions of project success, and a project of note, the interview moved into agile. Agile was introduced carefully: a way of working that has been getting increasing attention, but has not yet been extensively applied to the planning context. It was made clear that the way of working may or may not fit into that context- that parts may be more relevant than others, and that this research was talking to practitioners to explore that. Then came the last and longest section: getting practitioners’ thoughts on the agile lists from section 2.3. This part was really open-ended: the interviewees were asked to relate the characteristics, practices, and barriers back to their work. They were encouraged to elaborate on whatever thoughts came to mind- this was where connections between agile and planning were explicitly made. Sometimes comments were positive, sometimes negative, and other times confused.

Other best practices were followed during the interviews, such as making markers on relevant items to probe at a later point (different probe types explained by Whiting (2008, p. 38)) and concluding in a professional and friendly manner. In an effort to have consistency across sessions, the researcher created a checklist of items to address during the introduction and conclusion sections.

(26)

After the interviews, audio was transcribed and thematic analysis was performed in line with 11

the phases of Braun & Clarke (2006) (see Figure 9). Transcripts were coded in Atlas.ti, and coding was a combination of inductive and theoretical (Braun & Clarke, 2006). Potential themes were identified from the broad list of codes generated. Due to time constraints, only

the first 6 interviews were transcribed to reach the beginning of Guest, Bunce, & Johnson’s 12

(2006) 6-12 interview saturation point range.

Figure 9: Phases of thematic analysis (Source: Braun & Clarke, 2006, p. 87)

After generating the initial themes (which begun as semantic, but progressed towards

interpretative in certain places) from the transcriptions, the audio of the rest of the interviews 13

was listened to and coded. Themes were reviewed and revised, and then all of the data was re-coded before a final determination of themes was made. Relationships to the final two

sub-questions are explained in the Answering sub-question 2 and Answering sub-question 3

sections. For this method’s contribution to sub-question 1 (targeted question- How accessible

and relatable is agile to planning departments from the practitioners' perspective?),

responses from the Exploring one way of working item and general conclusions from the

researcher were used. This is described further in section 4.1.2.

3.2.3 Assisted process analysis (sub-question 2)

The third component of the research was an “assisted process analysis”- an adaptation of the project management method of business process analysis ⁠(Project Management Institute, 2013; Scottish Qualifications Authority, 2007; Swanson, 2013) to the social science context of

this research. This method looked at sub-question 2 (What are gaps between the current way

of working and agile in planning departments?) through the lens of a specific project of the Bicycle Program of the Municipality of Amsterdam while the interviews addressed the

question more broadly (Figure 10). First, the method will be explained in more detail, and then the case of the specific project analyzed will be introduced.

11_{There were challenges to this that will be explored in the reflection section}

12_{There were 11 audio recordings in total (1 interviewee preferred not to be recorded)- the transcripts}

and recordings have been securely stored and can be verified upon request

(27)

Figure 10: Sub-question 2 addressed through different lenses

The assisted process analysis is a hybrid method that consisted of three parts: an initial search for online content on the project, narrative interviews (Jovchelovitch & Bauer, 2000; Muylaert, Sarubbi Jr, Gallo, Neto, & Reis, 2014), and process mapping (Biazzo, 2002; Damelio, 2011; White & Cicmil, 2016) exercises. The content search was done individually by the

researcher, while the latter two were done together during sessions with two project stakeholders- one within the Municipality of Amsterdam and one from outside of it. These different components were used to look at the project from multiple angles. Figure 11 shows how the different parts come together for the analysis.

Figure 11: Assisted process analysis workflow

Online content search

The first step of this method was an online search for general information on the project in order to ground the research in its context. This was a surface-level exploration to get a general idea of what happened from publicly-available online information. It was a way to both complement the stakeholder sessions and to prepare for them. The following choices were made beforehand: the researcher chose to look for content related to what happened, the process, and the way of working for the project. Eyes were kept open for articles, social media posts, and any other miscellaneous web pages. What was found was organized and a timeline was made to estimate what happened based on these sources.

Narrative interview & process mapping sessions

The narrative interview and process mapping parts took place during sessions with two project stakeholders (for their privacy, results have been anonymized). Each component is explained individually first, and then how they are done together is described.

(28)

A narrative interview consists of initialization, main narration, and questioning phases (and at the end, concluding “small-talk”) (Jovchelovitch & Bauer, 2000). Essentially, the interviewee is probed with an initial question (in this case prepared as “Can you tell me the story of the Alexanderplein project?”) before proceeding to tell their story. The researcher’s role is to listen and not interfere with the narration. While listening, he or she makes markers on items to probe about later that are related to the research questions (i.e. about the process and way of working in the project). Once the interviewee has told his or her story, the researcher asks follow-up questions from the markers noted down. Narrative interviews allow going “beyond the transmission of information or content, making the experience revealed… they allow the deepening of research” (Muylaert, Sarubbi Jr, Gallo, Neto, & Reis, 2014).

Process mapping involves “constructing a model that shows the relationships between the activities, people, data and objects involved in the production of a specified output” (Biazzo, 2002, p. 42). A process map “show[s] clearly what a system does, what controls it, what it acts on, what means it uses to perform its functions and, what it produces” (Biazzo, 2002, p. 46). Its use comes through providing insights on how a given process may be improved (Biazzo, 2002). For this session, a swimlane diagram (Damelio, 2011) (example in Figure 12) was chosen as the type of process map for participants to create because it shows a workflow across different entities or stakeholders, allowing for insight into how the different

stakeholders worked together.

Figure 12: Example of a “swimlane” process map diagram (Source: Damelio, 2011, p. 7)

The first half of the session was a narrative interview where the researcher heard about the stakeholders’ personal experiences with the project. After the narrative portion of the session was complete, a brief explanation of process mapping was given. A fact sheet (see appendix) that was prepared beforehand was also given to the to participant to read for a few minutes to give them an idea of what they were going to be doing. Then came the exercise, where the participants mapped their perceived stages of the working process undertaken in the intervention. During the mapping exercise, the researcher was there to answer questions about process mapping, but the participant was given the lead on the exercise and it was them that made the diagram. As this exercise was a bit of an experiment by the researcher, time was left at the end to reflect with the participant on the exercise and its limitations.

(29)

Analysis

After the sessions, the two narrative interviews were transcribed and thematic analysis was done in line with Braun & Clarke (2006). Like for the analysis of the semi-structured

interviews, the researcher generated a broad list of initial codes (coding was partly inductive and partly theoretical) and identified potential themes. Then the initial themes were

determined, reviewed and revised. The themes started as semantic but moved towards interpretative during consolidation and revision. Data was reviewed and re-coded before a determination of final themes. Findings from the thematic analysis were combined with what was learned from the process maps and online content search to answer the targeted portion

of sub-question 2: What are gaps between the current way of working and agile for a specific

planning project?

(Sub-)case: Alexanderplein

The specific planning project selected for this research was the turning-off and removal of the traffic lights at the Alexanderplein intersection in Amsterdam (Glaser, 2017) (Figure⁠ 13). This was chosen because it was a significant recent project of the Bicycle Program of the Municipality of Amsterdam and because the author was already somewhat familiar with the intervention from involvement in as a student volunteer for research on the intervention in 2016.

Figure 13: Alexanderplein’s location within Amsterdam (Map data by Google)

After consulting many stakeholders, the Municipality of Amsterdam decided to do a pilot project on Alexanderplein where it turned off the traffic lights to see if flow was improved at the intersection. Research studies accompanied this trial to see how it was going, including one of intercept interviews from the University of Amsterdam that studied the human experience of people cycling through before vs. after the intervention. Despite people’s nerves, the trial was declared a success and the intervention was extended for a few months. Eventually, it became permanent and the intersection was fully redesigned. Now, the

Alexanderplein intervention is one of several interventions the municipality has taken to improve flow of cyclists.

(30)

3.2.4 Guided questionnaire sessions (sub-question 3)

The fourth and final component of the research was guided questionnaire sessions. It

addressed sub-question 3 (Under what conditions might agile be adopted by planning

departments in practice?) in a targeted way by testing two concepts as potential solutions to enable an agile way of working in practice. It was also a quantitative complement to the interviews, which focused on barriers in a general and qualitative way (Figure 14).

Figure 14: Different parts of sub-question 3 targeted with guided questionnaire sessions and interviews

The strengths and weaknesses of the two concepts were used as a means to explore the conditions under which agile could work in planning practice. The concepts selected were intended to test, not prescribe, how a potential solution might look like in practice. Due to the exploratory nature of ideas behind in the questionnaire, among other reasons, the researcher accompanied the participants in sessions instead of distributing the questionnaire for them to do remotely. First the content of the questionnaire will be described, and then format of sessions will be explained.

Questionnaire design

The questionnaire (Bryman, 2012, chapter 10) was comprised of 3 sections. The first section 14

asked participants to rate each item from the Barriers to the adoption of an agile way of

working list (Table 5 in section 2.3) on scale of 1 (not a barrier) to 5 (major barrier) in the context of a planning department. After the scale questions, participants were given space to explain how they would propose to overcome barriers that received a 5 (major barrier) and to write thoughts they had that could not be expressed in the scale questions.

The second and third sections tested potential “solutions” to enable an agile way of working

in practice: a pattern language for cycling (te Brömmelstroet, Nello-Deakin, Quillien, & 15

Bhattacharya, 2018)⁠ and the bicycle user experience concept (Hahn, 2016). The pattern 16

language was selected because of its potential to build a shared understanding between planners and communities of solutions that work for citizens. The bicycle user experience

14_{The full questionnaire can be found in the appendix}

15_{A pattern language is “a grouping of related patterns that work together”, and “an individual pattern…}

[is] a honed solution or configuration which successfully resolves the conflicting forces in a recurring context... a stable solution” (te Brömmelstroet, Nello-Deakin, Quillien, & Bhattacharya, 2018, p. 5)⁠.

16_{While it is called bicycle user experience, its application of human-centered design principles and}

(31)

concept was chosen because its human-centered design methods claim to produce infrastructure more in line with people’s needs. These are both in line with this research’s main question, which looks at how to deliver projects that better serve citizens.

Each potential solution had its own section, and within each section there were two parts. The

first part asked participants to rate each item from the Characteristics of an agile organization

(Table 3 in section 2.3) and Practices of an agile organization (Table 4 in section 2.3) lists on

how much the solution could enable it in practice. The second part asked how the solution

would deal with each item from the Barriers to the adoption of an agile way of working list

(Table 5 in section 2.3). Scales of 1 to 5 were used, and a does not apply (N/A) option was also given in case the respondent did not think the agile characteristics or practices connected to the solution (which was to be expected as this was exploratory and the connections between these concepts and agile weren’t known beforehand). At the end of each part, an optional open question was given for participants to record thoughts that were not captured in the rating bubbles (free-text comments are a valuable potential data source, as described by Rich, Chojenta, & Loxton (2013)).

In line with best practices for structuring questionnaires (Krosnick & Presser, 2010), the easiest questions were placed at the beginning of the questionnaire (i.e. the barriers list that

participants had already addressed qualitatively during the semi-structured interview). Additionally, sections were clearly defined and the structure was communicated at the beginning of the session.

Guided sessions

The questionnaire was administered in guided sessions for multiple reasons. First, completion of it required a basic understanding of a pattern language and bicycle user experience. Introducing them to the participants in person made it easier to make sure participants had a basic understanding of these. Second, the language around agile had potential to be

confusing- both conceptually and in translation. In-person sessions are more flexible and better suited for participants interviewing in their second language (i.e. English instead of Dutch), as mentioned by Barriball & While (1994). Third, as this was on exploratory on a conceptual level, it was not clear how participants would respond to the questions. By being present, the researcher was able to both clarify any of the participants’ doubts that arose and to take notes on their spur of the moment thoughts and reactions to the exercise.

The sessions were conducted with three practitioners from inside the Municipality of Amsterdam that had already participated in the semi-structured interview portion of this research. A practical understanding of the planning department was critical here in order to give feedback on where these solutions offer potential and where they fall short in real-world conditions. Thus, participants fell into two categories: planners/designers and senior staff (major decision makers). The planners/designers work on projects on a daily basis, while the senior staff works on a larger organizational level- together, there is a good understanding of the organization and the daily practices. Two of the participants fell into the former category, and one fell into the latter.

(32)

The session was structured as follows. First, an introduction set expectations for the session, showed participants the agenda, and explained that the researcher was present to clarify any doubts that arose. For each section, the sequence was as follows: the topic was introduced, reading material was given, and clarity of the material confirmed with the participant before the completion of the questionnaire section commenced. Reading material was pre-selected

by the researcher to be as concise as possible: the Barriers to the adoption of an agile way of

working list (Table 5 in section 2.3) for the first section, the most relevant parts of te Brömmelstroet et al. (2018) for pattern language and Hahn (2016) for the bicycle user experience concept (the latter with some additional descriptions added).

Analysis

Analysis was for this section was quantitative and qualitative. For the quantitative

questionnaire results, responses were averaged out and interpreted by the researcher. On the qualitative side, free-text comments from the questionnaire and the researcher’s session notes were interpreted. Additionally, findings from the first section of the questionnaire on general barriers to an agile way of working were used to supplement the interview targeted

portion (section 4.3.2) of sub-question 3 (Which barriers are most relevant to the adoption of

agile by planning departments in practice?).

3.3 Triangulation

Throughout the design of this research, the concept of triangulation (Carter, Bryant-Lukosius, DiCenso, Blythe, & Neville, 2014; Mathison, 1988; Nolan & Behi, 1995; Thurmond, 2001) was considered. Namely, attempts have been made at methodological triangulation by

supplementing the semi-structured interviews with additional methods to go deeper into the sub-questions, and at data triangulation by collecting diverse types of data. Additionally, for the assisted process analysis, sessions were held with stakeholders both inside and outside the Municipality of Amsterdam, and the combination of narrative interviews and process mapping exercises aimed to broaden what was learned. The goal of all of this was to approach the research (sub-)question(s) from multiple angles and reach deeper analysis.

(33)

“The customer is very wide… who is the

customer? You are a different customer

than I am. And people in other branches

over there. We all are different customers.”

“A program differs from a project

in that it focuses on targets

instead of results. Therefore, we

are flexible, adaptive in the way

we deal with projects...”

“Sometimes you need- you know- to step out a little bit out of your comfort zone in order to advance as a team.”

“Very important, it all comes down

to time. Our major concern is to find

time windows, time opportunities to

realize our projects.”

“For our group it is kind

of important, that there

is a good amount of

money available... For

the bigger projects I

would say money is not

an issue.”

customer at all times. And you have to

really take him- take his hand- through the

process... And I also think you cannot get

their attention early enough...”

“Yeah. It’s always important. There’s

never enough money for things.”

“I like to give people

a lot of responsibility.

And I also like to have

autonomy...”

“Information is very important. We are used to base our positions on- on information, but also on proven information.”

and interactions

over process and

tools… I think that’s

what it’s all about.”

CHAPTER 4

(34)

4. Findings

4.1 Sub-question 1:

How accessible and relatable is agile to planning departments?

4.1.1 Literature review

The literature review gave an academic perspective on how accessible and relatable agile is to planning departments. First, each source is discussed, and then implications for the (targeted) sub-question are explained. The findings are summarized in Table 10.

Context of planning department Agile Vonk &

Geertman, 2008

• The PSS implementation gap provides a lot to learn about the context of planning departments

• Items have been identified to better fit PSS into practice (see Figure 15)

• The authors are participating in (at least advocating for) a discussion with planning department stakeholders and PSS

developers about the real issues that exist around making PSS work in practice. • A common way forward that works for all stakeholders is necessary instead of just pushing technology on planning

departments

• Agile needs to have a conversation with the planning context, find common interests, and work together to make it happen in practice

Te

Brömmelst roet, 2012

• There is a disconnect between the people creating the PSS and the actual receivers/users

• Certain traits are important in deploying something in a planning department (i.e. Simplicity, transparency, flexibility, communication)

• Improving PSS implementation will be a mutual learning process (also alluded to by Vonk & Geertman (2008)), which

necessitates a shift in how the PSS developer acts

• Values behind agile connect well to what is predicted to be well-received by planning departments, but it’s critical that agile’s packaging and implementation mirror these values

• It may be good to refocus on the end goal (i.e. organization functioning better) and reframe the conversation if “agile” is perceived as a negative buzzword

Nerur & Balijepally, 2007

• The context of the planning department and the “wicked problems” it faces are linked to the intellectual thought behind agile

• Agile is part of a wider trend of thinking that transcends the software

development buzzword

• Characteristics of agile link back to theories and metaphors from several different contexts

Chan & Thong, 2009

• General organizational barriers to agile can be reflected on for the context of planning departments

• Acceptance of an agile methodology rests on knowledge management outcomes, the methodology’s

characteristics, and technology related factors (based on their conceptual framework- see Figure 17)

• Knowledge management outcomes come from ability, motivation, and