Data as Infrastructure
for Smart Cities
Orchestrating data marketplaces using the
SMARTify Dynamic Business Models Framework
This is a storyboard of the Ph.D. Thesis: "Data as Infrastructure for Smart Cities"
How to cite the content of this page and the Ph.D. thesis: Suzuki, L.C.S.R. (2015) "Data as Infrastructure for Smart Cities", PhD Thesis, University College London'.
If you want to know more about the framework or have any question about the contents of this page, please do contact the author:
firstname.lastname@example.org - www.larissasuzuki.com - @LariRomualdo
Copyright: Dr. Larissa Romualdo Suzuki 2015 ©.
The content of this page can be reused if citing the source.
Every day nearly 180,000 people move to cities, creating more than 60 million new urban dwellers every year. Alongside population change, cities physical infrastructure is struggling to cope with the increasing demands placed on it. Inadequate infrastructures have often been accompanied by the aggravation of many challenges associated urban living in terms of law and order, health, utilities, education, transportation, and delivery basic public services.
Cities have to plan for population growth, and introduce a more sustainable, efficient, and liveable model in urban development. Digital technologies offer a new wave of opportunities to create a balance between social, environmental and economic opportunities that will be delivered through smart city planning, design, and construction.
To enter into the new era of cross-domain data integration cities must to adopt a more strategic and outcomes-oriented approach. One that will facilitate the publication, management and dissemination of public and proprietary data, while also addressing privacy and trust issues in relation to citizen's volunteered data. These processes are supported by a data infrastructure.
City data is an incredibly important asset and is the foundation upon which smart cities vision will be built. Providing efficient access to public and private sector and citizen's data has the potential to enhance and transform both government and businesses services, as well as stimulate innovation in city services to the benefit of everyone.
Data infrastructure is defined as
"the basic physical, digital, organisational and governance structures and processes needed for the management of all data that underpins the decision making processes in smart cities" (Suzuki, 2015).
DATA INFRASTRUCTURES FOR SMART CITIES
Cities across the globe are demanding data infrastructures which can effectively capture, communicate, orchestrate, store, access and share city data gathered from the physical world.
City data is produced by a multitude of systems, devices and applications, and whose logistical distribution varies according to its suppliers (citizens, public and private data providers), their sectors, its distribution channels, and the policies and regulations to which it is subjected to.
The lack of understanding of the combined impacts that technology, stakeholder requirements, and big data will have on the overall operation of the city environment has resulted in the creation of many isolated data catalogues and "(self-denominated) platforms" which are not meeting the needs of their users. As a result, each initiative is composed by numerous fragmented data sources, and different technological standards and approaches.
The provision of cross domain city data can be facilitated through the design of an intelligent data infrastructure based on the middle-out leadership pattern defined in the SMARTify framework. It takes social influence into account while maximizing the efforts of other stakeholders who are working towards the achievement of the same goal: to create better data infrastructure which will unlock the data that realise smart cities.
In developing such principles, the implementation of data infrastructure does not require cities to start from scratch. A one-size-fits-all approach to data infrastructure transformation and simplistic approaches to engage stakeholders of city data with one another are unlikely to work.
TYPES OF City Data
The systems that operate city infrastructure have increasingly being tied to a pool of multi-structured real-time data which is catalysed by millions of electronic networked devices responsible to manage and operate the city infrastructure (e.g. sensors, smart meters, cameras, and actuators). Such environment made of connected things has created an entirely new dynamic network of networks known as the Internet of Things (IoT) which has revolutionized the world with the idea of the connectivity of anything, from anyone, at any time, and anywhere. Data collected by different sensors, actuators and devices is usually diverse in nature (temperature, light, logs, multimedia, etc.), mostly location and time dependent, and present different levels of quality. The diversity, instability, and ubiquity make the task of processing, integrating, and interpreting the real world data a challenging task. As it is often difficult to understand the context associated with sensory data, it is very hard for stakeholders and machines to access and interpret data unambiguously.
Government Open Data
More recently, governments around the world have provided open to thousands of government open datasets to offer additional data needed to address complex urban problems. Government Open Data initiatives have released vast amount of data (transport, water, environment, geospatial) for public access. The key fundamentals of Web 2.0 technologies include the idea of reusable, open data and open interfaces (APIs), so that data from one service can be combined with data from another to create interesting data combination and integration.
Citizens will be responsible for a great part of the huge volume of data expected to arise in London. This data will mostly come from weblogs, social media, mobile devices, automobiles, smart cards, smart homes (Home Area Network - HAN), and others. Humans are very powerful devices in themselves and in most cases have sensing, actuating, and computing capabilities that go well beyond pervasive devices currently available yet to be invented. For sensing, it is a matter of fact that there are situations and events that only human sensitivity and experience can recognize while pervasive devices have limited capability for such tasks (e.g. quality, interpretation of complicated contexts). To enhance sensing and computing capabilities of our urban environments, users should play an active role, and should contribute their own sensing and computing devices (e.g. cars, mobile phones).
These types of data can be offered as
Open data: non-privacy-restricted and non-confidential data. Produced with either public or private resource. Made available without any restrictions on its usage or distribution.
Private data: restricted and/or licensed data including permission, privacy, publication and distribution; as well as data that is presently held privately merely because it has not as yet been recognised to offer value. Produced with either public or private resource.
Commercial data: licensed data including permission, charging, use and distribution. Produced with either public or private resource.
The need for
Despite the considerable potential of city data, the technical, strategic and organizational issues make it difficult to capture their potential opportunities. There are many challenges associated with the current fragmented logistical distribution of city data. Some of these challenges are data interoperability, lack of strong platform leadership and strategies that enables open and proprietary data providers to co-exist and co-operate, and the disregard of new data business models, regulations, policies, use and re-use.
Data platforms and portals around the world face a number of difficult challenges to coordinate public, private and crowd-sourced static and real-time city data. These include:
• City-wide fragmentation in the logistical distribution of city data;
• Provision of several obsolete, non-valid, and non-value-adding data;
• Substantial human workload to clean them up for machine processing and to make them comprehensible;
• Lack of widely-accepted standards for expressing the syntax and semantics of city data;
• Non definition of policies, licences and regulations for city data re-use and commercial exploitation;
• Non-realistic scale ambition for the realisation of a "data platforms" (e.g. changes not occurring in small increments);
• The wide range of stakeholders and delivery partners involved in the delivery of a data infrastructure;
• Managing the tensions and conflicting expectations between the providers and consumers of city data.
Taken together, these challenges mean that both isolated top-down and bottom-up data management approaches cannot work. Success cannot be delivered by planning all the elements of the data strategy in an isolated fashion. The stakeholders of city data have different requirements and expectations with regards to the city data, and the dispersion of city data provided by different organisations results in the adoption of non-interoperable standards and technology agreements.
The current city data supply chain may have once upon a time have worked but is today too simple and expensive to desing and maintain.
THE SMARTIFY FRAMEWORK IN A NUTSHELL
The SMARTify Framework can be seen schematically in the Figure below. This approach combines both a top-down (government) and a bottom-up approach, the latter of which is emerging from communities advancing innovation and industry creating new standards. This new form of "middle-out" design when coordinated by efficient governance strategies can guarantee successful leverage of a data infrastructure. This middle out approach, which is formed on the basis of social influence and not authority, is formed by the following domain areas:
This framework identifies five key domain areas which assist on identifying the activities and processes needed to manage city data, and maximizing the efforts of the stakeholders who are working towards creating smart cities.
These domains are modelled and assessed during the life-cycle of data infrastructures which we define are: Research and Development, Procurement, Roll-Out / Implementation, and Market. The data infrastructure design is likely to be affected by specific external forces in each phase of its life-cycle. The data infrastructure life-cycle, external forces and the components of the five domains of the framework will be presented in another research paper.
The five components of the SMARTify Framework are:
1) SERVICE: Smarter services design to deliver integrated value propositions
Provide tailor-made data services which careful targeting and needs of users and businesses, and explore use cases where data is used to deliver different forms of value.
2) TECHNOLOGY: Smarter technology design for widespread exploitation of data
Create agreements between stakeholders regarding data handling and technical infrastructure to allow the design of infrastructures that will serve as the foundation for widespread exploitation of data in the long-term.
3) ORGANISATION: Smarter Value Network Design to maximize the efforts of stakeholders
A statement of values which city leaders can use to steer a strong value network of collaborators who will provide the expertise needed to deliver a data infrastructure. The value network take social influence into account while maximizing the efforts of other stakeholders who are working towards the achievement of the same goal.
4) VALUE: Smarter Value Design to explore new and innovative business models
Cities must implement new and transformational business models that are made possible by increased access to data and closer integration between city systems, and to change existing processes in order to capitalize on these.
5) GOVERNANCE: Smarter Governance to create impact and accelerate smart cities growth
Cities must ensure that the intended benefits of a smart city and a city data market strategy are clearly articulated, measured, managed, delivered and evaluated in practice. The city needs to deliver practical guidance on how to address city-wide challenges of joining-up across city silos and orchestrating their data marketplace.
Embracing the technology and non-technology components of data infrastructures will ensure that standards are adhered to, interoperability is guaranteed, smart governance is in place, a strong value network of partners are built, and feedback is facilitated. The employment of this framework can balance complexity and quality deliver a foundation for widespread exploitation of data in the long-term.
Smarter services design to deliver integrated value propositions
The change of powers in city data offering are beginning to happen whether cities plan for them or not, driven by the increasing adoption of social media, technology adoption, widespread of mobile services, and by and by rising engagement of citizens with public services and their expectations on the degree of interactivity they want from city services (e.g. mobility application).
The success of the data infrastructure and its services will be determined by mainly determined by the users' perceived value . Many existing city data platforms/portals raise new challenges in security and privacy since users implicitly expect their data to be secure and privacy-preserved. Developing business models and enabling individuals and businesses to discover new and inventive ways to use city data is also crucial to the large-scale dissemination and impact of services.
For this, cities will need to join forces and seek partnerships with both government and private sector so that there are resources available to enable data and services to be provided, maintained and enhanced to citizens. Cities must focus to deliver new digital services that will address the city and societal needs of cities in a positive manner. In order to accomplish this the city must engage with citizens and businesses as owners of and participants in the creation and delivery of city data and digital services, not as outsiders who are merely passive recipients of such services.
Ensuring societal needs for city data is recognized as the starting point for city data service offering can be a powerful driver of data service transformation. However, significant barriers need to be addressed.
My research and case studies have shown that in order to achieve a large-scale adoption and impact cities should:
• Implement a clear and comprehensible policies and regulations around the data being provided by the data infrastructure, which encompasses security on the delivery of data and privacy protection of all volunteered citizens data;
• Provide multiple channels for data release, and different levels of data access is offered to citizens, public and private sectors according to the terms and conditions to which the data is associated to;
• Implement mechanisms that ensures data services complies with relevant National or International regulations, government open data policies and legal agreements with regards to commercial data;
• Provide easy-to-use tools which facilitates data findability, discoverability and the traceability of the provenance and ownership of data;
• Promote engagement activities with the public to help individuals and businesses to learn and use the services provided by the data infrastructure;
• Create partnerships with external organizations that will provide the data and services needed on a city-wide scale;
• Prioritize sets of data that are in greater demand by city data consumers, and use them for the trial of standards and new technologies, and to explore and promote use cases where data is used to deliver different forms of value;
• Agree on a set of principles that the providers of city data can commit to in order to ensure the validity, quality, interoperability, trustworthiness and value of the data being offered and used in the data infrastructure.
Smarter Technology Design to create a foundation to the widespread exploitation of data
The technology design for the city should be based on a citizen-centric, interoperable, open and innovative vision. This implies a significant change on the operational model of the current city digital strategy, so that stakeholders can collaborate, disperse data can be reused, and fragmented silos of digital assets and services can be brought together to deliver smarter services.
To accomplish those changes, cities need to understand the existing technical and non-technical barriers of data management. On one hand cities must ensure systems and data complies with technical and semantic standards aimed to address interoperability issues. On the other hand, cities must tackle several non-technology aspects hindering the effective interoperability of systems and data, including different strategies to manage personal data, to exchange data among different stakeholders, and licenses terminology.
Data infrastructures design involves shifting the existing data repositories away from a silo-based delivery of data and digital services towards an integrated, scalable, multi-channel, and engaging service delivery approach: an approach that enables cross-domain city data and enables decision-makers to take advantage of unprecedented insights into how the city and its infrastructure functions.
For this, smart cities will need to adopt measures to address issues from both technology and non-technology perspectives. My research and case studies have shown that cities should:
• Define a technical architecture which is simple enough to be comprehensible at least at a high level of abstraction;
• Design a modular based architecture which relies on stable and well-defined interfaces to ensure interoperability between the platform, services and the applications provided by services complementors;
• Analyse the architectural features of data and its usage by performing a city data mapping exercise to develop a picture of the architectural features of city data, including: sources, volume, variety, temporal factors and sensitivity;
• Explore the vulnerability aspects of city data (e.g. volunteered citizens data), and define data management strategies which ensures data integrity and compliance with National, European and International data protection regulations;
• Ensure the data infrastructure is evolvable and able to accommodate additional functionality at later stage at a fair and transparent cost;
• Identify integrated approaches to design and service delivery which ensures that appropriate standards are adopted and put in place so that that services and data fit together and that synergies can be exploited (e.g. Standards Hub, IEEE P2413, W3C Semantic Web Standards).
Smarter Value Network Design to maximize the efforts of stakeholders
City data and digital services have been for a long time delivered by centralised but often dispersed and diverse set of city portals, developed by different organisations, using their own standards and technology agreements. As a result, the supply chain of data and services is almost exclusively composed by a platform provider, limited data providers and the consumers of city data.
This city data supply chain may have once upon a time have worked but is today too simple and expensive. Today's simple reality is that cities find it tremendously difficult to specialise in all the competencies involved in designing, building and maintaining an intelligent data infrastructure. Even powerful organizations like Google and Apple need to collaborate with the various members of their value networks (e.g. developers) in order to provide unique and inventive services and applications to end users. Hence, relationships collaborate for increased trust and commitment within the platform and are an important aspect of stakeholders collaboration in value networks.
In a data-driven value network (DDVN), processes and data are integrated in orchestrated supply chains which enable collaboration and co-opetition (the notion of collaborating in some parts and competing in a healthy way in other parts) in the provision of city data and services, as well as ensuring a response to demand that creates innovation and value.
There are many different stakeholders associated with the development of data infrastructures, each one with their own expectations, objectives, requirements levels of commitment and influence on the data infrastructure. Basically, there are three basic types of partners in a DDVN: structural partners, contributing partners and supporting partners with varying degrees of power within the value network, based on their resources and capabilities.
• Structural partners: have the most power in the network and are often formed by the platform providers and advisory partners. They actively promote the strategy and impacts of the data infrastructure. It will include the city authority, local elected representative, investors, strategic partners and regulatory bodies.
• Supporting partners: are the providers of services, tools and data, individuals and organizations involved in the delivery and the early stages of the data infrastructure, and validators of the strategies. It includes open and proprietary data providers, application developers, academic institutions, local councils, business partners.
• Contributing partners: are the direct users of the data infrastructure, creators of business cases, providers of feedback, and creators of knowledge and insights from the city data. It includes end users, data integrator, data scientists, application developers and the media.
This complex network of stakeholders can potentially bring insights about specialised domains and different application markets that one single organization or city government developing a data infrastructure for smart cities would struggle to maintain in-house. Basically, this concept drives cities to deeply specialise in their core competence – governance. As a consequence cities are able to decentralise their city data portals and platforms to create specialised supply chain networks which involve many partners, forming a large ecosystem of expert collaborators. SMARTify highlight that the success of data infrastructures will be co-determined by the way the value network is managed and nurtured.
In order to manage a data-driven value network, my research and case studies have shown that cities should:
• Disclosure technical and architecture blueprint details in order share and outsource expertise, and partnerships, and integrate supporting partners’ solutions into the infrastructure itself;
• Develop the data infrastructure in an iterative and collaborative manner, which includes structural and supporting partners and contributing partners involvement through research and digital engagement (e.g. social media, online tools which facilitates public participation in the process);
• Ensure our strategy is aligned and fits into the wider smart city’s future, and the city’s core strategic objectives;
• Provide the necessary resources, information and attention to the stakeholders to ensure a their ongoing participation on the development of the infrastructure, and to have a formal managed stakeholders engagement program in place.
Smarter Value Design to explore new and innovative business models
The value of city data can be divided into two categories: Value from Releasing data and Value from using data.
• Value from supplying data: The value created by realising urban data is not only in monetary and economic terms, but also from society and transparent governance perspectives. Actuators, sensors, mobile phones, smart cards, smart meters, open government data and social media generate vast amounts of structured and unstructured data. Within all this information lie many potentially profitable insights regarding modelling city services performance, spatial aspects of the city, land usage, citizens mobility and travel behaviours, and trends.
• Value from consuming data: Users are able to integrate data from a variety of sectors and distribute it across different domains, value chains and stakeholders, thereby supporting intelligent decisions in the urban environment and the creation of valuable new businesses through integrated services.
Understanding the different values that can be created through the use of urban data is essential to identify the enablers and the type of data necessary to unlock a specific value. For instance, monetary and good governance values can be unlocked through the release of aggregated data, while innovative services such as apps and new business require a more granular level data that is real/near-real time and with good quality. Competitive advantage is originated from innovative value-added services on top of data, and providing opportunities for innovation and the creation of new businesses through integrated data. On the other hand, urban data in human-readable format may empower citizens to make better decisions in their lives (e.g. crime rates, gas emissions, teachers per student in city schools) and increase their participation in public affairs, which in turn may raise the level of public trust and perceived responsiveness of government actions.
A fair division of costs, revenues and investments is required to make the collaboration worthwhile for all the stakeholders in the DDVN. In addition, the revenue mechanism for customers and/or end-users needs to be defined, alongside costs (including transaction costs), pricing and performance indicators.
My research and case studies have shown that cities should:
• Explore effective ways to recover costs of opening up data by seeking investment sources and creating alliances with the public and private sectors;
• Support diverse business models for the open and commercial exploitation of city data (e.g. subscriptions). Charging users to consume data may increase competitiveness on the platform, in which data providers will compete in providing the best quality data to end users;
• Demonstrate the impacts and value of city data through its independent use or re-use or by combining city data with other data sources. The use and re-use of city data can generate a range of benefits, which in turn can create value for data consumers and providers alike (e.g. economic growth, and better policy making using city datasets that improve value for money and the efficacy of policy).
Smarter Governance to create impact and accelerate smart cities growth
Cities need to establish leadership strategies which enable technology, data, digital assets to be managed as city-wide resources. While architecture can reduce structural complexity, governance can reduce behavioural complexity. Data Infrastructure owners must shape and influence its ecosystem of data and stakeholders, not to direct it, besides respecting the autonomy of complementors while also being able to integrate their varied contributions into a harmonious whole.
It is the platform governance that determines whether innovation divisibility made possible by data infrastructure is successfully leveraged. The current increase in the sources of city data and connected devices can exacerbates the complexity of the management of a data infrastructure and become a potential for inefficiency, duplication and unreliability. Platform openness and governance has the power to overcome the main challenges hindering the success of data infrastructures: which are regulations and policies, data services provision, data integration and misuse.
The level of openness indicates the degree to which new stakeholders can join the value network and are allowed to drive the strategy or provide services on the infrastructure (for example). The higher the desired level of control and exclusiveness is, the more likely a closed model will be adopted for the data infrastructure. On the other hand, reaching many stakeholders may be an argument in favour of choosing an open model. Cooperation between stakeholders and institutions is very important to develop and to maintain an efficiency data infrastructure for all. Active cooperation can shift the emphasis from the power to decide to the power to transform (i.e., deliver), key to overcoming the delivery deficit of efficient urban services .
In data infrastructures, the orchestration of it should be from a city and citizens strategic point which at the same time carefully takes into account the needs of the stakeholders and technology in the city.
In order to implement data infrastructures, it is necessary to have a good understanding of the platforms target markets, including similar urban data platforms initiatives, industry and users needs, trends, and cost structures.
Establishing and maintaining both competitive and collaborative relationships with collaborators will ensure that jointly activities which is based on commonly negotiated terms and conditions can be carried out. The goal behind many partner agreements is the optimization of the data infrastructure operations and services. By entering these agreements we can directly benefit of our partner’s or supplier’s economies of scale or of it’s specialized knowledge, which cities could not achieve on their own. Decision makers and providers of data infrastructures should reflect on what kind of partner resources could leverage their business model and their own competencies.
My research and case studies have shown that to put governance arrangements around data infrastructures cities should:
• Establish clear governance and accountability arrangements to ensure active participation, management of key risks and issues, compliance with defined rules, and successful delivery of outcomes;
• Implement program management activities to track the progress of the data infrastructure development, performance and feedback;
• Include public participation and consultation as an active and ongoing process throughout the development of the data infrastructure;
• Ensure the strategy is clearly articulated and it is aligned to deliver a clear and compelling vision which supports the cities' own smart city vision;
• Select key partners to develop practical, citywide, and fair commercial models to allow useful and interoperable data to be made widely available;
• Assist the owners of both public and private data to agree on a criteria as which data generated should be made available and explore use cases where data is used to deliver different forms of value.
University College London - Computer Science
www.larissasuzuki.com - @LariRomualdo
University / Supervision
University Collge London / Professor Anthony Finkelstein - @profserious