A Strategic Approach for implementing A Smart Pedestrian Network (SPN) System

1 Introduction

Software development for large scale applications must take into account the long term and adopt a strategic approach to systems analysis and design. In addition, emphasis should be given on project management and implementation issues considering application in a variety of contexts. Especially, if we aim at developing systems for smart cities which involve changing end-user behaviour, it is imperative that implementation issues should be given great attention. For successful system development and implementation we need to bring together multiple stake-holders, so that we integrate views and gain acceptance. In this context, software development becomes a co-creation process that could borrow techniques from Open Innovation, Customer Involvement, and co-creation [3, 11]. Such a software development process case is the Smart Pedestrian Network System (SPN) [14, 15, 16]. In this way, SPN software should consider aspects of urban planning and development. Specifically, the SPN system should be analyzed on the social, economic and environmental levels.

In the case of SPN, a clear problem statement is necessary so that the extent of its domain is properly analysed. The SPN system development can then be justified with the fact that Information and Communication Technology (ICT) can act as a catalyst for promoting active mobility. Clearly active mobility could alleviate problems with air pollution. It is estimated that there are 4.2 million deaths each year directly attributed to air pollution [20]. With over half a million premature deaths in Europe it is evident that there is a huge problem to deal with. As air pollution mainly stems from high levels of vehicle use alternative modes of transport such as walking and cycling would have a significant impact in tackling the situation. Further, current ecosystems and ocean life are also being adversely affected by CO2 emissions and the resulting climate change. Via a Smart Pedestrian Network system we can develop and promote sustainable transport modes in order to reduce pollution and global warming. In this way, we can significantly mitigate the climate change crisis by maintaining cleaner air in our cities, preserving our natural environment and land, and therefore, conserve the planet for ourselves and future generations.

In the current transport model used in our cities, we see that we heavily rely on vehicle use and current Intelligent Transportation Systems (ITS) solutions mainly support motorised vehicles [10]. Sustainable urban transport policies can be enhanced with the use of ICT to substitute vehicle use with active means of transport. On the other hand, this is not an easy solution to implement as people still opt to take the presumed more convenient and easier route, which is using a motorized vehicle instead of engaging in active mobility. This can be attributed to lack of awareness of appropriate walking routes, as well as the lack of knowledge of the many benefits that active mobility can have both to ourselves, and to our planet. There is a lot to be gained if we embark on more active means of transport, such as simply walking. With walking, we can have a plethora of benefits for ourselves and the environment. The key to this sustainable mobility transformation, is not only to encourage and convince, but also support people when engaging in active mobility. All of these can be supported by the provision of relevant information.

Current advancements in ICT and Artificial Intelligence (AI), create a significant opportunity to support the active mobility vision [2]. For example, a smartphone application system could be developed so that people are informed about appropriate walking routes, but also give feedback via crowd-sourcing, to municipal authorities on issues regarding the pedestrian network conditions. Even though several applications may give general information on the pedestrian network environment, they are not necessarily geared towards suggesting appropriate and enjoyable walking routes, and thereby encouraging people to walk more. Also, they do not provide information on safer routes, and are unable to inform users about amenities such as water fountains, public toilets or resting places that certain user groups such as elderly people may particularly need.

Even though SPN presents an innovative proposal, that on its own does not guarantee a successful implementation. It is important to consider the use of an appropriate framework for properly implementing SPN. As shown later in the literature review, current methods for software development, which include the Waterfall Model and the Agile Model, do not adequately address these implementation issues. We investigate such issues and suggest a road map for successfully implementing our SPN system.

In the next section we examine past literature and software development methods and the potential for mobility navigation systems. We also discuss the effectiveness of methods to support the successful development of software systems.

2 Methods

According to research studies, as of 2018 [12], a majority of smartphone owners use a navigation application to help them in tracking a suitable route for their journey. The Manifest study [9] surveyed more than 500 smartphone owners to understand how people are relying on technology more to get from point A to point B. Their findings were as follows:

1. Over three-fourths (77%) of smartphone owners regularly use navigation apps.

2. Google Maps (67%) is the most popular navigation app by a wide margin. It is the preferred navigation app for nearly 6x more people compared to the second-most popular app Waze (12%).

3. One-quarter (25%) of smartphone owners say they use their preferred navigation app because they believe it offers better directions.

4. Over one-third (36%) of smartphone owners use navigation apps prior to leaving their location.

5. 34% use them en route, and 30% use them both prior to leaving and en route equally.

6. An overwhelming number of respondents (87%) use navigation apps for driving directions.

7. People use navigation apps for non-driving directions such as public transportation, walking, or cycling twice as much in the Northeast than in other U.S. regions. This might be due to cultural reasons and better infrastructure conditions.

From the above, we can derive several implications for the development of navigation systems suitable for pedestrian use. The fact that an overwhelming 87% of respondents use navigation applications primarily for driving directions, is a disturbing fact. This fully supports our argument that there is an extremely high level of vehicle use in our cities. Furthermore, it provides evidence that vehicle use is proving to be a major contributing factor to environmental detriment and the deterioration of human health.

As a result, there is a lot of potential for people to use navigation systems for active transportation such as walking and cycling. This showcases that indeed not only is there demand for such products, but that this demand could be enhanced further if an appropriate smartphone app is developed that appeals to the public. Hence, there is a potential market granting our SPN app the opportunity to achieve its goal in promoting higher walking levels and reducing air pollution.

Such findings may suggest that the reason so many individuals use navigation applications for vehicle use, is because there currently does not exist in the market a satisfactory application designed specifically to promote sustainable mobility. Even though some applications have been developed for this purpose, these failed to take off and become widely used. Table 1 provides an analysis of the main applications found in Google Play. Existing applications fail to completely satisfy particular pedestrian user needs such as getting information on amenities, safe routes, comfortable mobility etc., when engaging in active transportation.

Further, in a market research and concept study for SPN [13], it was shown that there is demand for such a smartphone pedestrian network app so that people are informed about the various aspects of the pedestrian network including comfortable mobility, safety, convenience, and connectivity. Furthermore, it was found that successfully implementing SPN would be challenging, as high levels of investment and commitment from the side of the municipal authorities are required. Therefore, new business models that involve the element of sponsorship and co-creation should be considered.

Therefore, what is important to also stress in this section, is that for the SPN system to be a success, it must be properly implemented using an appropriate framework. This is discussed in the following paragraphs by reviewing the most widely used methodologies for developing information systems. These include the waterfall approach and the agile method.

The waterfall methodology model [6] is a sequential software development process where progress flows steadily toward the conclusion - like a waterfall - through the phases of a project. That is, analysis, design, development and testing. The nature of the Waterfall methodology insists that each phase be completed and perfected before the start of the next phase. This involves fully documenting a project in advance, including the user interface, user requirements, and all the features’ variations and outcomes.

A detailed investigation and full research into a product feature are conducted up front, eliminating (most) project risks. With the bulk of the research done in advance, estimates of the time required for each stage are more accurate, thus providing a more predictable release date.

During the analysis stage, the product development team analyses the requirements, and fully understands the problems. This is a research phase that includes no building. The team attempts to ask all the questions and secure all the answers they need to build the product requirements. In the design stage, the software developers create a technical solution to the problems set out by the product requirements, including scenarios, layouts and data models. This phase is usually accompanied by documentation for each requirement, which enables other members of the team to review it for validation. Once the design is approved, technical implementation begins. This is often the shortest phase because research and design have been done in advance. Upon completion of full implementation, testing needs to occur before the product can be released to customers. The software testing team will use the design documents, personal and user case scenarios delivered by the product manager in order to create their test cases.

Very early on, Royce [17] in a study about managing development of large software systems criticized the sequential nature of the waterfall method. It was asserted that software should not be developed like an automobile on an assembly line, in which each piece is added in sequential phases. In such sequential phases, every phase of the project must be completed before the next phase can begin. There are criticisms against this approach in which developers first gather all of a project’s requirements, then complete all of its architecture and design, then write all of the code, and so on. Specifically, this approach has been opposed due to the lack of communication between the specialized groups that complete each phase of work.

This is where the agile method comes in to fill in the gap, as an improved method of software development. According to McConnell [6], the agile methodology is an iterative approach to software development that is performed in a collaborative environment by self-organizing teams which produce the final product version in a sequence of multiple product releases. The methodology produces high-quality software in a cost-effective and timely manner to meet stakeholders’ changing needs.

Every product release begins with a list called a “backlog”, which consists of a list of prioritized requirements which are dynamic and can change at any time. Determining the next product release, a smaller number of requirements can be chosen from the backlog list. The work involved in the next product release is called “iteration (or sprint)”. Each iteration has a short time frame which usually lasts from one to four weeks, depending on the team’s experience. A key element of an iteration, is that the priorities regarding which requirements should be built, do not change within the iteration; this list should only change from one iteration to another.

The Agile methodology accepts that project change is inevitable. The use of small iterations allows changes to be absorbed quickly without inflicting significant project risk. The system to be developed can be adjusted as business priorities evolve. With the next iteration, the software development team can adapt to the new priorities.

It is easy to see how the “waterfall” methodology is far from optimizing the final software solution compared to the agile methodology. First of all, it assumes that every requirement of the project can be identified before any design or coding occurs. But in truth, this is seldom the case. This is an impossible task that makes the waterfall model inadequate for effectively developing software solutions. On the other hand, the waterfall model provides structure to software development which is essential for complex projects such as SPN.

Therefore, a combination of the waterfall and agile methodologies would be ideal for the case of developing the SPN information system and smartphone app. We have attempted to follow the waterfall methodology in outlining a roadmap in order to analyse the SPN structure, and future goals we hope to achieve within a given period of time. This was also done for the SPN design, implementation and testing phases. In order to add flexibility and effectiveness in developing SPN, we incorporated the dynamic approach as suggested by the agile methodology model. Therefore, an integrated approach that combines the waterfall and agile models is proposed as shown in Figure 1. Note that the approach involves multiple iterations with elements of co-creation and open innovation.

Figure 1

Integrating the Waterfall and Agile methods with elements of Co-Creation and Open Innovation

Further, we see that current methods for developing software interactive systems, such as SPN fail to effectively consider implementation issues. As we see further on in this paper, we explain how the proposed approach can deal with such matters, which involves making the system ready for the market via testing, adaptation, feedback and adjustments, to improve the chances of successful implementation.

3 Results

We are currently in the process of developing the SPN system which provides real time analysis of the actual physical and virtual environment that exists in a specific area within the pedestrian network. Furthermore, the SPN system assists pedestrians in formulating their interests, preferences or goals at any given moment in time regarding walking routes, specific attributes, objects of interest and their properties. SPN is able to calculate alternative routes based on pedestrian preferences, topography, and actual status of the objects in the physical and virtual environment in real time and present a holistic vision of the route options to pedestrians on-demand. Additionally, it records walking routes, measures relevant statistics, collects updates regarding actual status and properties of route objects, and presents it to pedestrians.

SPN offers re-routing upon calculation of (unfavorable) changes in the status of particular objects along the route and deviations from the initial route plan and schedule by time, distance, physical condition, health, degree of fatigue, etc. It has also been devised to support final assessment of route results by the pedestrian. This is achieved by collecting important data for statistics, analytics, prediction and feedback. Finally, SPN could process the results and update statistics regarding the particular city condition, and pedestrian experience based on personal profiles.

3.5 The Proposed Strategic Implementation Framework for Scaling Up

The proposed framework for implementing SPN is based on an integration of the waterfall model and the agile methodology. It is comprised of two phases. Phase 1 presents aspects of implementing SPN in one particular city, while Phase 2, examines the issues of implementing SPN in multiple cities. In Figure 2 which depicts Phase 1, we show the very beginning of implementing the SPN system. To secure funding and support from municipalities and investors, it is necessary to properly present our business plan. Explanations should be given to our potential stakeholders, why SPN is worth investing in, how it will benefit society and our environment, and how we will get people interested in using SPN. Also, explanations should be given on how the budget is prepared and utilized. Once having attained the necessary support and funding, part of it, would be allocated for all the necessary tools, equipment and technology that is needed to make SPN successful.

Figure 2

Roadmap Phase 1

Once an initial version of the SPN system is agreed upon with stakeholders, the next step is to test it. Testing is to be carried out with the cooperation of the municipal authorities in order to adapt SPN to the context and needs of the particular city. Municipal authorities and a group of end users would be required to use the system for a given number of weeks, after which they will provide their feedback. We can use any positive feedback to further promote SPN. For any shortcomings, we will make the necessary adjustments and changes to satisfy particular needs of end users and municipal authorities.

Having had first-hand experience with implementing SPN in Phase 1, we will proceed to Phase 2 as depicted in Figure 3. In Phase 2, the SPN system is ready to be launched in the multiple municipalities that are interested. Priority would be given to municipalities in various European countries. We could begin with the countries of the associated organizations/institutions, which are part of the SPN partnership. These countries are: Austria, Cyprus, Italy and Portugal. In this way, we can better monitor the SPN system’s performance.

Figure 3

Roadmap Phase 3

All necessary data and feedback from stakeholders should be collected. This will be used to assess various factors including:

1. how frequently the SPN app is being used on average;

2. SPN application performance;

3. which markets are using the SPN app more;

4. number of weekly/monthly downloads, ratings;

5. if the market is increasing via widespread use;

and other factors that deem necessary for monitoring a successful SPN implementation.

This information is to be used in order to notify governments and municipalities about various issues regarding the pedestrian network. Successful collaboration with governments and municipal authorities is essential in order to invest in improving infrastructural conditions, walking paths, providing amenities etc. It is important to consider feedback from stakeholders as this will help in improving the SPN system. Feedback will be taken into consideration and any necessary changes or adjustments will be made to meet and satisfy users’ needs.

4 Discussion

This paper presents our experiences from developing a smart pedestrian network (SPN) system, which aims at promoting walkability in urban spaces. Even though the system considers effectively the variety of user needs, that does not guarantee its successful implementation. Going through a literature review on the main methods of developing information systems, we found that there are gaps when it comes to supporting the implementation of interactive information systems, such as the SPN case. As a result, we have developed an implementation framework for SPN that would also be useful for other similar interactive systems development. The proposed framework integrates aspects of the waterfall model and the agile software development method.

Especially by means of prototyping, valuable information may be collected ensuring that the system to be developed is in accordance to end user needs. In addition, stakeholders’ views could be taken into consideration with the use of an appropriate prototyping method. Moreover, a co-creation approach should be implemented as in the case of SPN, where developers, stakeholders and end users will meet in organized workshops to share their views on the current prototype. Such views and opinions could then easily be integrated in an improved prototype until the final system development.

Further, the need to take a multi-stakeholder approach in software development becomes even more important nowadays. In the current critical situation with the COVID-19 pandemic, ICT and software development take a new, prominent role in our lives. Urban interactive systems development, such as the case of SPN could be further developed into integrated social smart city systems to manage the COVID-19 crisis. However, software development becomes a very complex process, that requires new tools and approaches, such as the ones presented in this paper.

Finally, software development should make use of ideas and concepts of open innovation. Open innovation principles would enhance the requirements engineering process but also system analysis and design. In the case of SPN, such an approach is implemented to emphasize testing, adaptability and feedback in order to give room for adjustments to be made based on future demands.