Easyfleet – A Customised Fleet
Management
- A case study in developing a web
application that supports navigation,
simplicity and customer satisfaction
Jonathan Axelsson
Viktor Cheng
Renata Felippe da Silva Galeano
Eric Hermansson
David Lindholm
Simon Olefalk
Cecilia Shadman
Supervisor: Dennis Persson Examiner: Aseel Berglund
Copyright
The publishers will keep this document online on the Internet – or its possible replacement – for a period of 25 years starting from the date of publication barring exceptional circumstances.
The online availability of the document implies permanent permission for anyone to read, to download, or to print out single copies for his/hers own use and to use it unchanged for non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility.
According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement.
For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its www home page: http://www.ep.liu.se/.
© Jonathan Axelsson, Viktor Cheng, Renata Felippe da Silva Galeano, Eric Hermansson, David Lindholm, Simon Olefalk, Cecilia Shadman
Abstract
More vehicles are used in as part of business operations and the need for obtaining an overview of a vehicle fleet is increasing. This study sets out to examine whether or not it is possible to implement a web based application to manage a vehicle fleet that promotes navigation, simplicity and customer satisfaction, with the Finnish Border Guard as the customer. Theory regarding design, navigation, simplicity in web development, usability and customer satisfaction testing was studied. A prototype was initially formed to provide the customer with expectations of the project. During the implementation phase, user tests were conducted in order to measure to what extent the purpose of the project had been fulfilled. The user tests were partially used as a basis to further implementation and eventually as the final evaluation of the project. The final study shows that improved navigation can be achieved by having a static and constantly reachable navigation bar and a minimalistic design can conduce to simplicity, although it is unclear whether you can generalise that fact.
Table of contents
1. Introduction ... 1 1.1 Explanatory statement ... 1 1.2 Purpose ... 1 1.3 Question formulation ... 1 1.4 Limitations ... 1 2. Background ... 3 3. Theory ... 43.1 Cooperation and Communication ... 4
3.2 Customer satisfaction ... 4 3.3 Design ... 5 3.4 Usability ... 7 3.4.1 Navigation ... 7 3.4.2 Simplicity ... 9 3.5 Method-theory ... 10
3.5.1 Agile working methodology ... 10
3.5.2 NABC ... 11
3.5.3 Prototyping ... 11
3.5.4 Customer Satisfaction testing ... 12
3.5.5 Usability Test ... 15 3.5.6 Evaluation scales ... 19 4. Method ... 20 4.1 Pre-study ... 20 4.1.1 Marketing plan ... 20 4.1.2 Product backlog ... 20 4.1.3 Prototype ... 21 4.2 Implementation ... 21 4.3 Evaluation ... 22
4.3.1 Customer satisfaction testing ... 22
4.3.2 Usability testing ... 23
5. Results ... 26
5.1 Pre-study ... 26
5.1.3 Prototype ... 27 5.2 Implementation ... 28 5.2.1 Technical overview ... 28 5.2.2 System overview ... 29 5.2.3 Database structure ... 36 5.3 Evaluation ... 36 5.3.1 Usability testing ... 36
5.3.2 Customer satisfaction test ... 38
6. Discussion ... 40 6.1 Result ... 40 6.1.1 Pre-study ... 40 6.1.2 Implementation ... 40 6.1.3 Evaluation ... 44 6.2 Method ... 46 6.2.1 Pre-study ... 46 6.2.2 Implementation ... 47 6.2.3 Evaluation ... 47 6.2.4 Source criticism ... 49
6.3 The project in a broader context ... 49
6.3.1 Ethical aspects ... 49
6.3.2 Societal aspects ... 50
7. Conclusion ... 51
8. References ... 53
9. Appendix ... 57
Appendix A - Marketing plan ... 57
Appendix B - NABC ... 72
Appendix C - Prototype ... 74
Appendix D - ER-diagram ... 79
Appendix E - User tests ... 80
1. Introduction
When vehicles are an important part of companies as part of their business
operations, it arises a need for obtaining an overview of their vehicle fleet. Storing this information using old-fashioned means such as spreadsheets or whiteboards is time-consuming, inefficient and not accessible at all time.
In this day and age, a web application could not only fulfill existing needs but also provide new and valuable functionality. The main advantage of storing information about a vehicle fleet in a web application would be to provide easy access to information for everyone in the organisation at all time.
1.1 Explanatory statement
A couple of years ago, the US Coast Guard developed a web application for the same purpose. The Swedish Coast Guard learned about the existence of their system and determined that it could be useful for them as well and subsequently developed their own version. Their Finnish counterparts, the Finnish Border Guard (hereafter referred to as FBG) have similar needs and a similar vehicle fleet. However, they do not currently possess a similar system.
With this knowledge, the project team initiated a discussion with FBG about
introducing this kind of system. There was considerable interest and willingness to participate from their side, which led to an agreement between the team and FBG. This concluded in an agreement in which this project will have the purpose of
developing a web application that can provide a better overview of their vehicle fleet. The web application was designed with usability and more specifically with navigation and simplicity in mind in order to minimize the resistance to change to a new solution.
1.2 Purpose
The purpose of this project is to develop a web application that can help the FBG store and display information about their vehicle fleet and make this information more accessible within their organisation.
1.3 Question formulation
How can a web application with the purpose of improving vehicle management and interdepartmental cooperation be designed and implemented in a way that leads to high customer satisfaction and usability with regard to navigation and simplicity?
1.4 Limitations
This project has focused on the specific needs of the FBG and been developed in accordance with the requirements set by the FBG. While this type of web application
could be useful in other types of companies such as car rental companies, possibly with additional functionality, that was beyond the scope of the project with regards to the development of the web application.
Other limitations regarding FBG is that the application will be implemented on their intranet. The internal system will be maintenanced by an internal administrator. As to the customer satisfaction construct, dimensions that will be considered in this thesis are design, simplicity and navigation.
2. Background
Due to the fact that FBG was the customer for the project, the application had to be designed to fulfill their requirements. Through discussions with FBG, the following list of features was established as necessary to fulfill their requirements:
● Several users should be able to simultaneously access the system to view the availability of vessels.
● The system should contain 10-15 data elements (‘descriptors’) in order to accurately describe the current vessel (e.g. name, location etc.).
● Users should be notified in the web application when there are changes in the status of a vessel.
● Users should be notified via e-mail when vessels are marked as “not in use”. ● Support for uploading and attaching documents for each vessel.
● Every vessel should have a complete history log stored in the database. ● Each user should have an individual user profile and belong to one of the pre-
defined user classes with pre-defined user rights.
● There should be an administrator class with the ability to modify user profiles.
Beyond the external requirements from FBG, the project group was given free reins in regards to other functionality within the system. FBG did not express a wish for a purchasing process, which resulted in that particular functionality being removed in the final version that was delivered to FBG.
3. Theory
In this chapter, the theory the project is based on is presented. It starts off with the theory for the working method for the project. Thereafter continues with presenting the theory used for the three phases of the project;; pre-study, implementation and evaluation, chronologically.
3.1 Cooperation and Communication
For large companies, a management information system (MIS) is of great
importance. It is a system that satisfies communication and information needs of managers for organisations. Such a system concerns the process of collecting and storing relevant information amongst other tasks. The key focus of MIS is to
contribute to decision-making in a timely and accurate fashion at a minimum of cognitive and economic cost, for all managers at each level of an organisation (Felix, 2009). Managerial issues can widely be seen to pave the way for an absence of cooperation (Cheney & Dickson, 1982).
According to Felix (2009), a successful MIS contributes by:
● Providing warning signals.
● Mechanising routine operations in order to relax human work. ● Assisting managing functions with mundane decisions.
● Providing necessary information for non-mundane decisions. ● Gaining advantages on a market.
Communication and interdepartmental cooperation, in general, is a necessity for a successful organisation. To provide for prosperous interdepartmental cooperation, studies show that there is a need for high quality and credibility of the information that is being carried on between different departments within an organisation (Pinto & Pinto, 1990, p. 202).
3.2 Customer satisfaction
Customer satisfaction is a broad concept which could be described as the
satisfaction a customer get from being supplied with a product or service. Several factors have been recognised to affect the customer satisfaction on web applications, both regarding design and basic functionality of the application (Kassim & Asiah Abdullah, 2010) (Gao, 2005).
McKinney et al. (2002) argue that things that impinge customer satisfaction can be divided into system quality factors (SQ) and information quality factors (IQ), and the expectations regarding these quality factors. McKinney et al. (2002) further reason that there is no real consensus regarding what kind of dimensions of SQ and IQ that
are the most important, but they present five IQ dimensions: ● Relevance ● Timeliness ● Reliability ● Scope ● Perceived usefulness.
as well as four dimensions in accord with SQ: ● Access ● Usability ● Navigation ● Interactivity.
Nonetheless, there are several models to be considered while studying customer satisfaction on web applications, perceived simplicity of web design is an imperative parameter to consider (Gao, 2005). Other design dimensions to consider are design content, organisation, the basic structure of the site, where visual content is placed etc. (Kassim & Asiah Abdullah, 2009).
Moreover, there are several models, for example, TAM – Technology Acceptance Model, which evaluates the customer satisfaction primarily based on perceived ease of use. There seems to be a strong correlation between customer satisfaction and ease of use. As perceived ease of use is a somewhat loose term, it could be defined as the simplicity of the website with the level of user interaction and level of
understandable content as well as the level of navigability provided by the application (Straub et al. 2002) (McKinney et al., 2002). Gao (2005) also argues that perceived simplicity as a usability term is a factor which can positively affect the overall
customer satisfaction, although he means that there is a discrepancy between simplicity and ease of use.
The concepts of design and usability as navigation and simplicity will be further theorised below.
3.3 Design
User-Centered Design (hereafter referred to as UCD), which is a method that aims for a collaboration between the user and the designer, can be utilised to make the web application suitable for the user in terms of design. The design of the web application can be based on user information gathered from mapping their preferences and behaviour of use.
evaluation. During the first phase, the users and their needs are detected. During the design phase, the information from the design research is used to design the web application. During the third phase, the design is evaluated and eventually revised to better suit the user. Following these steps will, in theory, contribute positively to the usability of the web application (Williams, 2009).
Usage and the choice of colours affect the user’s overall experience of a web application. According to Issa & Isaias (2015) they attract the user’s attention and appeal the mind moreover they help users to understand information, memorize certain parts and acts as reminders. Different colours have various effects on the mind. Cold colours as blue, green and mixes of them give a feeling of seriousness, honesty and cleanliness. Hot colours such as red draw attention, and light colours are used to provide a feeling of openness. Light green, yellow and red are examples of colours that should be avoided using too much because they make the eyes tired as a result of the structure of the human eye. Mixing bright and light colours also makes the eye tired and is not to recommend (English et al. 2002). Other clean and tidy colour conventions, that support simplicity and are highly intuitive, are white, black and several shades of lighter grey (Kontratova & Goldfarb, 2007).
Though images can give a web application a richer visual environment, too many animations and flashing images can be distracting. Large pictures slow down the load time, on the other hand, users prefer graphical approaches on icons and buttons (Fang & Salvendy, 2003).
Fessenden (2017) means that modals were initially used to grab users attention in case of an error or a system issue that requires the user’s attention. Modals were accordingly used to interrupt the user to fix the error. Hence, the usage has developed to additionally grab user’s attention for less sincere reasons. When modals are considered to be put in place, there are both pros and cons to evaluate. Cons are that they cover information in the background which can make it harder for the user to respond to a question at the modal when the information needed is
covered. Modals also created an extra task for the user that can cut off the user if not containing important information. Though, Fessenden (2017) advises to use modals to simplify complicated processes that ask for a lot of information of the user which requires mental involvement. They attain the users’ focus on the specific matter and should therefore not contain what is not necessary for reaching the user’s goal.
When creating a form, the design of it matters in terms of simplicity. A form should not stand out too much when it comes to design, it should look clean and tidy. One column should be used and the fields and the text should be placed symmetrically. Eye-tracking experiments show that putting labels above the fields is preferable because it helps the eye to get a quick overview of the fields. If the form requires the user to fill in a date, only making it possible to do so manually should be avoided. Jarrett & Gaffney (2009) means this is because it can cause confusion when the user
has to relate the date to a specific time or happening, for example, tomorrow or the week before the holiday. To simplify date picking, a calendar that pops up when a date is to be picked should be used. Hence, the possibility to write in the date manually should always be a choice (Jarrett & Gaffney, 2009).
Users access a web application from different devices with different sizes of the screen. For the interface to be user-centered, the design should be responsive to match the user’s needs and the capability of the device. Techniques regarding how to make a design responsive can involve scaling the layout and changing the order of content on the page to prevent the user from scrolling too much (Majid et al. 2015).
The resolution of computer screens is in general low, which results in a higher risk of eye tiredness. Because of this reason and to prevent serious effects on a user, online typography is an issue specifically important to consider which includes the choice of font. There are two major types of fonts, the serif and the sans serif. The word serif refers to the short lines, or feet, that are attached to the edges of a letter or symbol. Serifs are due to their professional appearance suitable for printed media such as books or newspapers, but not as preferable for digital media due to the low resolution of screens which tends to clog the serifs together. Therefore, the sans serif is
recommended for text in web applications in order to make it simpler to read and attain information for the user (Kalbach, 2007).
When deciding where to place the company logotype on the web application, Whitenton (2016) means there are favourable advantages doing so at the top left corner. Additionally, by making the logotype clickable with the purpose of taking the user to the homepage simplifies the navigation for the user by the following reasons. Firstly, the user is reminded of its position at the web application and secondly, it provides an opportunity to whenever needed navigate to other parts of the web application or goes back to the start position.
3.4 Usability
Usability is a widely studied concept the in human-computer interaction field.
Researchers mean that it plays an important role in succeeding while implementing and designing computer systems and web applications. Poor usability in a web application can be linked to a broad variety of factors, e.g. navigational difficulties, inefficiency, difficulties in understanding the application’s content and disorientation. Upon identifying common dimensions of usability within a web application, several constructs are described, which seem to be the fundament of usability. Among these constructs, navigation and simplicity are included. (Lee & Kozar, 2012)
3.4.1 Navigation
navigation is of big importance. Poor navigation opportunities leave the user
confused, frustrated and lost which can result in the user leaving the web page (De Angeli et. al, 2006)
Leavitt & Schneiderman (2006) points out various focus areas to bear in mind when developing a web application. The conclusions are based on the information that the human eye navigates from the top to the bottom of the page, starting at the center and going to the left, lastly the right. Since the user is initially searching for important content at the top of the page, relevant information, buttons and functions should be placed at top of the page.
According to Nielsen & Loranger (2006), navigation within a web application refers to the way of letting a user access what is necessary for an efficient way. Consequently, in order to maintain a user-operational application, dynamic and customised
navigational items must be simple to manage. Keeping the site responsive, but not too responsive can be a key because over-sensitive functions and menus tend to become misleading (Nielsen & Loranger, 2006). Moreover, a well-recognised fact presented by Issa & Isaias (2015) is that a minimalistic edge design is cardinal to aid to navigation.
Additionally, studies show that an essential way of supporting navigation on a website is to have a constantly reachable navigation bar on each page, to decrease the likelihood of confusion (Issa & Isaias, 2015). To simplify navigation, Issa & Isaias (2015) and Nielsen & Loranger (2006) advocates using links that are labeled and placed properly. Furthermore, they mean that there should be a clear consistency in labeling, referring to link names, product names as well within searching mechanisms on a website.
One of the fundamental factors of overall navigation and usability is a search function. When a user scans a web application or website, they typically look for a placeholder where you can input words to navigate yourself to a requested layer of the application. Therefore, a placeholder as “Insert word or thing to find” plays a basal role in web application usability, especially in accord to navigation (Nielsen, 2012).
Upon creating a web application it is also fundamental to keep a consistent way of navigating. Other than keeping consistent labels, changing the navigational structure can lead to discomfort for the user. If the navigational structure changes from one page to the next, e.g. if you use search bars on one page and a filter function which performs the same actions on the next, the user’s attention drastically changes which can lead to dissatisfaction (Nielsen & Loranger, 2006).
A way of helping the user navigate on the page is by preventing mistakes from being made. User control and freedom is about helping the user leave a function when it has been chosen by mistake. This is usually done with an exit button or similar.
Errors can also be prevented by presenting confirmation options or by selecting information for users (Nielsen, 1994). In case that an error is made, it is also important that the user gets a clear response from the application of what had occurred and how the problem can be solved (Squires & Preece, 1999).
3.4.2 Simplicity
Nielsen defines simplicity as the lack of complexity, or in other words, the lack of obstruction. It is one of the key components when creating a usable design and having a web application with high simplicity means that users are able to get what they came for when they use the application (Karvonen, 2000). To reach simplicity in an application, there are four important factors: reduction, organisation, integration and prioritising. (Lee, Moon & Kim, 2007).
Reduction is the main factor behind simplification of a web application. The core of reduction is figuring what the essential parts of the application are. After that, the complexity of the application is reduced as much as possible while still maintaining it's essential functionality and design. When reducing, Waloszek(2004) concludes there are many different aspects of the complexity of applications that can be focused on: ● Terminological complexity ● Interface complexity ● Procedural complexity ● Navigational complexity.
The terminological complexity can be reduced by making sure the same names are used for the same functions and objects throughout the application. The reason behind interface complexity is usually bad screen design. Components can be badly arranged or misaligned. Correcting this helps to reduce the interface complexity. Reducing the procedural complexity is done by reducing the number of steps it takes to complete the different procedures in the application (Waloszek, 2004).
Organisation can be improved by grouping components into categories that are not abstract and makes sense to the user (Waloszek, 2004). When having a lot of
elements of the same kind, grouping them in a list also helps with the organisation of the application (Maeda, 2006).
When simplifying an application, an issue is often that the application ends up as a collection of isolated tasks and functionalities. To reach simplicity, these need to be integrated in an intuitive way. Using abstract such as a tree or net structures can lead to the application being perceived as more complex and thus less simple. Instead of using simple lists or graphical represents helps to reduce the perceived complexity (Waloszek, 2004).
Prioritising is the issue of having the application focus on its essential tasks. It is those which should be the most visible and accessible (Lee, Moon & Kim, 2007). This is connected to reduction since some functionalities with low priority should be discarded. Others should be marginalized so they don’t take focus from the
essentials (Waloszek, 2004).
Another way of influencing the user’s perception of simplicity is keeping consistency and standards when it comes to choice of words and expressions. To avoid
confusion the web application should use conventional words and expressions, that the user is already familiar with (Nielsen, 1994).
3.5 Method-theory
3.5.1 Agile working methodology
In software development, agile methodology is commonly used due to its framework that is “designed to help small, close-knit teams of people develop complex products” (Sims & Johnson, 2012). Agile methodology consists of various iterations, called sprints in scrum methodology (Sims & Johnson, 2012), and describes the different faces of the project. Iteration will hereafter be referred to as sprint. Figure 1 describes an iteration face, a sprint, and all its different steps (Otaduy & Diaz, 2017).
Figure 1, Agile methodology.
The process starts with defining a product backlog based on the features wanted in the final web application by the product owner (Sims & Johnson, 2012). The product owner can update the product backlog during the project to reflect for eg. changes in the needs of the customer (García-Magariño et. al., 2009). During the Sprint Planning stage, the team determines which features are going to be developed during each sprint and a sprint backlog is created. The next step of the process is the actual development of the features in the sprint backlog. These developed features are then sent to the customer for user acceptance testing. The feedback that the customer returns is then processed by the team and a list of changes is produced. These changes are then integrated into the product backlog. This marks the completion of one full sprint. Otaduy & Diaz (2017) state that a sprint should not be longer than 1 to 4 weeks depending on the size of the time frame.
3.5.2 NABC
According to Cristian (2015), NABC is a method used for the development,
assessment and presentation of an idea. It was developed by the Stanford Research Institute in the USA - originally for use in business situations - as a structured
approach to defining the value of an idea. NABC is used to pitch an idea, i.e. to try and convince others of the value of the idea.
The acronym NABC stands for:
N → Need, which describes the need on the market that the idea will satisfy. A → Approach, which describes the approach to satisfying the need.
B → Benefit, which aims to quantify the experienced benefit of the customer. C → Competition, which describes other competitors that can satisfy the need.
3.5.3 Prototyping
The process of prototyping is a significant part in expressing the design of a system or a product, and in mapping and exploring possible alternatives (Houde & Hill, 1997). According to Nielsen (2003), prototyping can also be used in usability testing, before the implementation stage of a project is started. By gathering user data to test the usability of the functionality and design before starting to code, it is possible to facilitate the programming phase.
Prototypes are usually classified according to their fidelity – high- or low-fidelity. Sauer, Franke & Ruettinger (2008) describe prototype fidelity as the degree to which the prototype resembles the finished system. A model, therefore, needs to show close to identical performance, functionality and design as the finished system, in order to be classified as a high-fidelity prototype. A low-fidelity prototype is on the other hand more of a cardboard example of what the finished system would look like (Blackler, 2009).
Advantages of low-fidelity prototypes are that they are cost effective and that it is easier to refine the design of the system. By using successive design refinements, the system can easily evaluate the usability of their system and functionality, although it might not be to the same extent as with a high-fidelity prototype (Rettig, 1994) (Rudd, Stern & Isensee, 1996). Sauer, Franke & Ruettinger (2008) claim that the limitations in recognising issues concerning usability come from the fact that a low-fidelity prototype enables different behaviour than a finished product.
High-fidelity prototypes enable making more realistic comparisons to actual designs (Blackler, 2009). However, high-fidelity prototypes are more time-consuming to manufacture as well as modify. As high-fidelity prototypes focus more on the
their focus (which is rather that of assessing the real functionality of the system). Assembling a high-fidelity prototype (e.g. an expensive functional physical prototype) before testing a low-fidelity prototype (e.g. a paper model) is also a matter of risk taking because a single manufacturing error in the high-fidelity prototype can make the evaluation stage impossible to complete (Rettig, 1994). This can be contrasted with the fact that a low-fidelity prototype can easily be remade, thus mitigating the effect of any errors and increasing the possibility of successfully completing the evaluation stage.
3.5.4 Customer Satisfaction testing
Several methods can be used when evaluating customer satisfaction. The most common ones include in-depth interview, evaluating the product/service with a focus group, telephone interview, mail survey and personal interview (Grigoroudis &
Siskos, 2010). Other tests and methods, such as usability tests can be used to gather information about customer satisfaction dimensions.
According to Grigoroudis & Siskos (2010) mail surveys can be an appropriate
solution in cases where directly contacting the customer is either too costly or difficult to perform. Among other advantages, Grigoroudis & Siskos (2010) mentions the following ones for conducting a mail survey to evaluate customer satisfaction:
● Respondents are not under pressure to provide quick answers. ● Different geographical areas may easily be covered.
● There is no interviewer bias.
The disadvantages mentioned by Grigoroudis & Siskos (2010) include:
● The questionnaire has to be short and questions should be simple. ● Questions can be skipped.
● Open-ended responses are more complicated to analyse.
Grigoroudis & Siskos (2010) covers six main steps that can be used to plan a survey. These steps consist of:
1. Determine survey objectives. 2. Determine satisfaction dimensions. 3. Determine measurement process.
4. Determine sample size and survey procedure. 5. Develop questionnaire.
6. Test questionnaire and refine.
discussed in more detail.
Grigoroudis & Siskos (2010) talks about that it is important that the questionnaire is made simple, with specific and single-minded questions and that the structure should help respondents when they are giving their answers. This can be summarized in with the KIS rule “keep it short, keep it simple, and single-minded” (Grigoroudis & Siskos 2010, p. 179).
Grigoroudis & Siskos (2010) argues that the survey typically contains four main sections:
● Introduction:
○ Information about the survey, why it is conducted, how the customers have been selected.
● Demographics:
○ Questions that are related to the demographic characteristics of the customer answering the survey.
● Satisfaction:
○ Questions about the how satisfied the customer is with parts of the product or with the whole product.
● Behavior:
○ Questions about the customer behavior and attitude towards the product or service.
Determine satisfaction dimensions
As discussed in the earlier section customer satisfaction is often related to other concepts. Grigoroudis & Siskos (2010) argue that product attributes, customer requirements, and performance among other factors also have an impact on customer satisfaction.
Grigoroudis & Siskos (2010) state that it is of high importance to determine relevant factors that might have relation with customer satisfaction. These factors clearly depend on the problem that the product/service is trying to solve.
Grigoroudis & Siskos (2010) mention one set of eight dimensions that can be used to measure customer satisfaction for general cases:
1. Performance (basic operating characteristics).
2. Features (secondary characteristics added to basic features). 3. Reliability (probability that product will operate over time).
4. Conformance (the degree to which a product’s design and operating characteristics meet established standards).
5. Durability (a measure of product life).
7. Aesthetics (subjective personal judgments regarding how a product looks, feels, sounds, tastes, or smells).
8. Perceived quality (general image of the company, reputation, and other subjective perceptions based on advertising, brand name, etc.).
Another dimension that is relevant to consider is how satisfaction has a relationship between expectations and result.
Grigoroudis & Siskos (2010) argue that a question about the overall satisfaction always should be included in the questionnaire. One definition of satisfaction is that it can be viewed as a function, equation 1, of an initial standard and some perceived discrepancy from the initial reference point (Oliver, 1980, p. 460).
𝑆𝑎𝑡𝑖𝑠𝑓𝑎𝑐𝑡𝑖𝑜𝑛 = 𝑓(𝑒𝑥𝑝𝑒𝑐𝑡𝑖𝑜𝑛, 𝑑𝑖𝑠𝑐𝑜𝑛𝑓𝑖𝑟𝑚𝑎𝑡𝑖𝑜𝑛) Equation 1: Satisfaction function.
Where disconfirmation is defined as follows: the outcome that has a worse result than expected in regard to the initial point is defined as a negative disconfirmation. Whereas outcome that is perceived better than expected is defined as a positive disconfirmation (Oliver 1980, p. 461).
When determining satisfaction dimensions that are to be used in the assessment process of customer satisfaction criteria one can consider the customer to be
deciding when evaluating a product or service. Therefore, the model presented below about identifying decision making can be used when defining the satisfaction
dimensions (Grigoroudis & Siskos 2010, p. 183).
Properties in modeling process of decision criteria:
1. Essential, to indicate consequences in terms of the fundamental reasons for interest in the decision situation.
2. Controllable, to address consequences that are influenced only by the choice of alternatives in the decision context.
3. Complete, to include all fundamental aspects of the consequences of the decision alternatives.
4. Measurable, to define objectives precisely and to specify the degrees to which objectives may be achieved.
5. Operational, to render the collection of information required for an analysis reasonable considering the time and effort available.
6. Decomposable, to allow the separate treatment of different objectives in the analysis.
7. Non-redundant, to avoid double-counting of possible consequences. 8. Concise, to reduce the number of objectives needed for the analysis of a
3.5.5 Usability Test
To evaluate and ensure that a product meets the requirements set by the customers and the developers, usability testing needs to be conducted. The overall goal of usability testing is to secure that the product is useful to and valued by the customer, is easy to learn, helps the customer to be effective and efficient at what they do and that the product is satisfying to use (Chisnell & Rubin, 2008). Dumas & Redish (1999) also states that the primary goal of a usability test is to improve the usability of a product. By eliminating problems with the design of products, frustration from using the product is reduced. This will also ensure that the customer will find the product useful, effective, efficient and satisfying (Chisnell & Rubin, 2008).
Although usability testing can result in a better product, there are still inherent
limitations associated with (usability) testing due to its artificial nature. Regardless of whether the test is conducted in a laboratory setting or at the actual site where the product will be used, there is no certainty that the test will resemble a real-life situation. This is partly because the act of conducting a test can itself influence the result. Another limitation is that it can be hard to test the product on the whole target population. This is because it can be difficult to identify every type, or the whole range, of possible end users (Chisnell & Rubin, 2008).
Chisnell & Rubin (2008) have defined three different tests that can be conducted during a product's development lifecycle, these are stated in the list below. The different tests are Exploratory Study, Assessment Test and Validation Test.
At the beginning of the lifecycle, regardless of whether it is a new product or a follow- up test on the previous release, an exploratory study is usually conducted. During this stage, the product is usually still being defined and designed. The objective is to examine how effective the product or the early preliminary design concept is. The exploratory study is conducted so that the developer or designer can determine to which degree the end user's intended usage goal is fulfilled and how well the navigation interface of the web application works. An exploratory study can also answer some of the developer’s questions about the user(s), such as how they
perceive the product and if the product’s basic functionality has the intended value for the user. The developer also receives a first impression of how well user(s) can navigate and which functions that might need to have written documentation to support their usage. (Chisnell & Rubin, 2008)
An assessment test is conducted after the fundamental product design has been established, which is in the beginning or in the middle of the product's development lifecycle. The main purpose of the test is to follow up on the exploratory study and evaluate the usability of the product. For example, if the exploratory study put the focus on page-navigation and design then the assessment test should focus on the functions supporting the front-end functionality and evaluate how effective the
implementation is. The exploratory study evaluates how intuitive the product is to use and the Assessment test examines how well the user can perform the intended task and identify usability deficiencies. Assessment tests are designed so that the user will perform tasks and the result will be quantified. (Chisnell & Rubin, 2008)
The validation test is the last test in a product's development lifecycle and often conducted in close proximity to the time of release. It is conducted to certify that established benchmarks or predetermined usability standards have been reached. It is also conducted to confirm that problems that were identified in earlier tests have been solved and to ensure that no new problems have been introduced. The
predetermined usability standards usually originate from usability objectives, which in turn are developed from earlier usability tests, interviews with users or educated guesses by the development team.
The usability objectives are usually focused on efficiency and effectiveness such as how effective the user can perform chosen tasks. The objectives can also focus on ranking or rating from users based on their preference criteria. An important objective of the validation test is to ensure that components work together with other
components in the final product as intended. (Chisnell & Rubin, 2008) Usability test design
When designing a test, it is important that the research questions are as measurable, accurate, precise and clear as possible. It is important to have a clear goal of what the test is supposed to evaluate or test. Without a clear goal or well-formulated
questions, the test might be a success but the outcome might not be usable (Chisnell & Rubin, 2008). Dumas & Redish (1999) states that the test participants should perform real tasks when conducting the usability test.
Chisnell & Rubin (2008) gives an example of two research questions that are too vague and unfocused. The example of questions is as follow:
● Is the current product usable?
● Is the product ready for release or does it need more work?
While the questions are understandable, they do not indicate how the results are to be quantified or measured. With this kind of question, there is a risk that those involved identifies different problems that some participants agree is a problem and some don't. This makes it hard to identify what problems need to be prioritised and dealt with. This also might lead to no problem identification at all (Chisnell & Rubin, 2008).
Following is a list containing a set of examples, defined by Chisnell & Rubin (2008), of product research questions that are suitable for websites or web applications:
● How easily do users understand what is clickable?
● How easily and successfully do users find the products or information they are looking for?
● Where in the site do users go to find Search? Why? ● How easily can users return to the home page?
● Can users perform common tasks within established benchmarks?
● How does ease-of-use compare between our product and the competition?
Research questions should be designed as close-ended questions. Scales,
statements or checkboxes will help to minimize the amount of responses that require extensive writing (Chisnell & Rubin, 2008).
A complement to a written form is the “concurrent thinking aloud” technique. The participants are asked to, while conducting the test, to actively tell the test moderator what they are thinking about while conducting the different tasks. This method helps to find tasks or problems with the product that makes the participant feel frustration or confused. Advantages with this technique are that it can give early indications on problem areas and help some participants to focus more on the test (Chisnell & Rubin, 2008). Other advantages are that it can result in a clearer overview of the problems that a user can encounter. (Van den Haak & De Jong, 2003)
Disadvantages are that the technique can be unnatural and distracting for some participants. The participants might also become exhausted before the test is over if the process goes on for too long (Chisnell & Rubin, 2008).
To ensure that the research questions are well designed, unambiguous and
unbiased, a pilot test can be conducted. A well-conducted pilot test will indicate how well the research questions are eliciting the sought for information. A pilot test should consist of the research questions plus a set of questions with the aim to evaluate the actual test. One example of such question is "Were the questions confusing?". It is also imperative to look for biased questions that might lead the test person to the correct answer. When designing research questions, it is easy to, unknowingly, design the questions in a way that leads the answers in "the right direction". The pilot test can also provide ideas on new questions or identify questions that are
redundant. (Chisnell & Rubin, 2008) Usability test participants
When choosing participants for a usability test it is important to know who the end users will be and choose participants that suit this description of characteristics. It is imperative to distinguish between the purchaser and the end user (Chisnell & Rubin, 2008). Participants should consist of end users or participants who represent end users (Dumas & Redish, 1999).
The number of participants will be important for the statistical validation of the test and according to Chisnell & Rubin (2008), 10 to 12 participants from a specific group of end users is minimum. This is also reinforced by Faulkner (2003). Chisnell & Rubin (2008) also state that in some, less formal usability tests, 4 to 5 participants who represent the same type of end user will find about 80 percent of the usability
deficiencies of a product related to that group of end users. Chisnell & Rubin (2008) are skeptical of involving only 4 or 5 participants because the risk is greater that a severe problem is overlooked. Faulkner (2003) states that at least 10 test participants are needed to ensure that 80 percent of the usability deficiencies are detected.
An important factor is that the number of participants will depend on the required degree of confidence in the result and the availability of required type of participant (Chisnell & Rubin, 2008).
Chisnell & Rubin (2008) defines a set of groups of participants that can be used for the study or test. Some of these groups are stated below:
● Internal participants:
○ Cheap and easy to find.
○ Can be biased because of affiliation to the company. ○ Good for testing the test.
○ Good for conducting an early exploratory study. ● Qualified friends or family:
○ Known in advance to be qualified.
○ Kan be biased but not as biased as internal participants. ● Website sign-up:
○ Good for when conducting extensive testing. ○ Motivated test subjects.
○ Might be biased because test participants are "fans of the company". ● Societies and associations:
○ Good participants who have a specific hobby or a specific profession. ○ Easy to reach through email-servers or forums.
● Referrals from personal networks, co-workers and other participants: ○ Good for finding participants from the whole age-scale.
● Existing customers from in-house lists:
○ Excellent source of experienced participants. Usability test analysis
When the test has been conducted, the data needs to be analysed (Dumas & Redish, 1999). As soon as the test has been conducted, a preliminary analysis is done so the worst, most urgent, problems are identified. This is done so that the developers can start to work on these problems while a more extensive analysis is conducted. The next step is to compile the data and summarize it. It can be beneficial
to start compiling the data during the test period to make this stage more effective. When summarizing the data, each question with a limited choice answer is
separately summed up to give a clear overview of trends. The next step is to analyse the data to get a better view of trends and patterns. Important during the analyse is to identify all the task that does not reach the benchmark goal and the tasks where the participant did not solve the task in an intended way (Chisnell & Rubin, 2008).
3.5.6 Evaluation scales
Grigoroudis & Siskos (2010) and Maurer & Pierce (1998) states that Likert scales are the most common one used in surveys. The Likert scale measures the agreement with a given statement, the scale consists of intervals that gives a degree of how much the participant agree within two adjectives of opposite meaning.
According to Grigoroudis & Siskos (2010), there is no consensus on how many options of answers that should be present in a scale, but they mention that earlier research indicates that the scales should not constitute of more than five options. If choosing an odd number of options in the scale, one should be aware of that the participant might interpret that there exists a neutral option which could be used to express indifference.
4. Method
In this chapter, the methodology of the project is presented. It is divided into the three phases of the project: pre-study, implementation and evaluation.
4.1 Pre-study
This segment is related to the first iteration in the project, sprint 0. Firstly, an idea for the end product was formed as well as the basis for the product backlog. Within the time frame for sprint 0, a marketing plan was constructed as well as an NABC analysis. Furthermore, a simplistic prototype was designed and presented to the customer.
4.1.1 Marketing plan
The marketing plan was initialised by a contemporary social and environmental analysis in order to examine the value of such a product on the market. Marketing strategies and goals were set based on the analysis. On a more detailed level, segmentation, customer targeting and market positioning were evaluated and set. Lastly, an examination of the ideal marketing mix was added so that pricing, product positioning, the value of the product and the right promotion was established. The idea of a fleet management web application was formed by having a conversation with the FBG. In addition to the requirement specification from the FBG, the project group aimed to add some more functionality to the web application. To develop and evaluate the idea, a NABC analysis was performed, aligned with the theories of Christian (2015). When proceeding with the analysis, a wider perspective of the market for the idea was considered. The idea was developed and evaluated towards general companies with a vehicle fleet, not only towards the requirements expressed by the FGB. The analysis helped the project group to assess the idea and identify functional aspects that the web application would include. Different aspects of the analysis were discussed within the project group, such as the needs of the market and the procedure in how to satisfy these needs, benefits with the idea and a competitor analysis. The discussion resulted in a number of key takeaways, which are presented in the result in this report.
4.1.2 Product backlog
The user story for the product backlog was initially based on ideas created by the group during a concept generation meeting in the early stages of the project. Later on, as the customer provided the specification of requirements, supplements to the user stories were added to the product backlog. All user stories were given a specific priority, in order to create a scheme during the implementation stage of the project, which can be related to the theory presented by Otaduy & Diaz (2017).
4.1.3 Prototype
The prototype that was compiled was based on the product backlog and the
specification of requirements that was provided by the customer. As the requirements were given at an early stage of the project, along with the fact the customer gave the project group space to develop other functionality quite freely, the focal point for the prototype that was formed was to showcase the functionality and basic layout. The prototype showed the structure of the web application without specific regards to the actual implementation of the system. This is a prototype that is in line with what Rudd, Stern & Isensee (1996) call a low-fidelity prototype, where the prototype only provides some basic visuals and workings of the finished product. An advantage with a low-fidelity prototype is that it has a high value in early design process stage
because it can easily be evaluated (Blackler, 2009) and the possibility of changing the design can be managed in early stages of the project. The main factors behind the decision to use a low-fidelity prototype were that it was less time-consuming and that it would facilitate the implementation stage.
Rudd, Stern & Isensee (1996) argue that a high-fidelity prototype can address navigational issues and it provides for realistic comparisons with other existing products. However, for this project, it would have been too time-consuming and the matter of presenting functionality proposals in time for the customer was integral. Therefore, a low-fidelity prototype consisting of wireframe mockups was created using the free software Balsamiq.
4.2 Implementation
The implementation stage of the project relates to the rest of the sprints, sprint 1, 2 and 3. Sprint 1 related to the first implementation stage, where core structures for the application were formed. During sprint 1, basic navigational functionality and a vessel list were implemented. Sprint 2 was the iteration where fine adjustments and special functionality, as well as more specific functionality, was implemented. The back-end structure was finished during this stage of the project. Sprint 3 was partly an
implementation-based iteration where finishing front-end touches were made.
The back-end of the application was built with Python and by using the web
framework Flask. Data was stored in a SQLite database and SQLAlchemy was used to access it. For the front-end, the standard techniques used were HTML, CSS and Javascript as well as two large frameworks to simplify the process, jquery and Bootstrap.
The database structure was designed by first creating an ER-diagram which included the most important and initially necessary elements and then continuously and
thoughtfully expanding it when new elements needed to be added throughout the project. The database was designed in such way that it would be simple and intuitive