=Paper=
{{Paper
|id=Vol-3223/paper14
|storemode=property
|title=A Reference Architecture for Participatory Budgeting in an eGovernment Landscape
|pdfUrl=https://ceur-ws.org/Vol-3223/paper14.pdf
|volume=Vol-3223
|authors=Achim Reiz,Michael Fellmann
|dblpUrl=https://dblp.org/rec/conf/bir/ReizF22
}}
==A Reference Architecture for Participatory Budgeting in an eGovernment Landscape==
https://ceur-ws.org/Vol-3223/paper14.pdf
A Reference Architecture for Participatory Budgeting in an
eGovernment Landscape
Achim Reiz1 and Michael Fellmann1
1
Rostock University, 18051 Rostock, Germany
Abstract
Participatory Budgeting (PB) puts a part of a city’s public budget directly in the responsibility
of its constituents – most notably, the citizens – who can make proposals on how to spend the
money and vote on the proposals they like best. This type of resource allocation can increase
democratic understanding and the efficiency of public spending. However, integrating or
adding such IT-supported PB applications into eGovernment systems is far from a trivial task
as a PB should not be isolated from the rest of the governmental IT systems but integrated.
Thus, the complex system landscapes of the municipalities have to be touched.
Reference architectures can guide the creation of future-proof software. Hence this paper is
meant to assist the implementation of PB into an existing eGovernment landscape by proposing
a PB reference architecture. It shall support administrations who plan to implement a PB by
proposing a possible system architecture and integration scenarios. As PB is part of an
eGovernment system, it also presents a state-of-the-art eGovernment IT architecture on which
other applications can build.
Keywords 1
Participatory Budgeting, Reference Architecture, eGovernment
1. Introduction
Participatory Budgeting (PB) puts a part of the public budget directly in the hands of the constituents.
The citizens can craft proposals for spending the budget, which are later voted on by the community,
with the proposals receiving the most votes getting implemented. This democratic process of allocating
parts of the budget increases the inclusion of otherwise overlooked groups, triggers debates on how to
spend the budget, increases the public’s understanding of governmental processes, and thus strengthens
the quality of democracy. Further, PB improves state performance by increasing accountability [1].
In recent years, the idea of PB spread worldwide and increasingly took hold in administrations
worldwide. Moreover, as PB initiatives often use specialized software ([2] provides an overview of
existing solutions), these systems often need to be integrated into the larger IT infrastructure landscape.
At best, these PB systems work seamlessly with other eGovernment (eGov) applications, sharing data
and logins. Such interoperable applications need software architecture management to abstract the
complexity to a controllable amount [3]. However, creating software architectures from scratch is not
a trivial task, as many systems need to interact, and the design choices directly influence the
corresponding systems. Reference architectures can support design decisions and mitigate the
complexities. They offer blueprints that can be adapted and integrated for individual use cases and
support the creation of high-quality concrete architectures [4].
While some research papers describe actual instantiations of eGov-architectures, neither a general
reference architecture for PB nor eGov systems are available. This lack leaves aspiring administrations
in need of such systems on their own. Even though their needs and challenges are often similar, every
administration needs to create its own IT architecture, which leads to unnecessary heterogeneity and
BIR 2022 Workshops and Doctoral Consortium co-located with 21th International Conference on Perspectives in Business Informatics
Research (BIR 2022)
EMAIL: achim.reiz@uni-rostock.de (A. 1); michael.fellmann@uni-rostock.de (A. 2)
ORCID: 0000-0003-1446-9670 (A. 1); 0000-0003-0593-4956 (A. 2)
©️ 2020 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
152
costs. Our research aims to fill this gap. This paper’s goal is to derive a technical reference architecture
for these PB applications into a broader governmental IT landscape. At first, we identified already
existing PB initiatives and gathered the state-of-the-art in eGovernment architectures based on a
systematic literature analysis. In the next step, the core features of a PB were integrated into a universal
service-oriented eGov architecture. While the current focus of the architecture is on participatory
systems, we believe that the architecture can also be used to integrate other kinds of eGov systems.
2. LITERATURE SEARCH
A systematic literature analysis first gathered the current state-of-the-art. The analysis is grounded in
the methodology for systematic literature reviews in information systems research proposed by Barbara
Kitchenham [5]. The search focuses on the main research question: “Which architectures were
developed that are relevant for the IT support of PBs?” Papers were included as relevant if they
presented an ICT-related architecture for PB or related fields (#1) or architecture where a PB can be
integrated (#2). Thus, we initially queried for PB-specific architectures (#1.1, #1.2) and software
architectures that are similar regarding their overall expected components (#1.3). Later, we included a
more general view of eGovernment architectures (#2.1, 2.2) and integrated the PB-related aspects into
the eGovernment view. We excluded architectures that are domain-specific without relevance for PB,
literature reviews, or such that are unreadable (e.g., due to low resolution). PB is an increasingly global
phenomenon. Thus, researchers from different regions and continents published results. The search was
performed using the metasearch engine Scopus, which offers extensive coverage of peer-reviewed
research output from various sources.
Table 1
Development of the Search String for the Systematic Literature Analysis
# Search String Results Relevant
1.1 TITLE-ABS-KEY((“participatory budget*” OR pb) architecture AND 24 –
reference)
1.2 TITLE-ABS-KEY(("participatory budget*") AND architecture) 9 [6]
1.3 TITLE-ABS-KEY(("participatory budget*" OR "idea management" OR 125 [7]
"innovation management") AND architecture)
2.1 TITLE-ABS-KEY((egov* OR e-gov*) reference architecture) 89 [8,9]
2.2 TITLE-ABS-KEY((egov* OR e-gov*) AND architecture AND framework) 440 [10–19]
3. DEVELOPING THE REFERENCE ARCHITECTURE
The identified architectures differentiate the core elements like user interfaces, application servers, or
data storage. While some research, like [10] and [14], do not further organize the underlying
architecture, most of the other papers ([8,11–13,15–19]) propose a layered structure. This kind of
architecture can encapsulate various levels of separated functionality, with each higher layer using the
functionality provided by the lower levels. Layered structures allow for easier standardization due to
the abstract definition of standardized interfaces and tasks [3]. The layered architectural structure is
thus also inherited for the newly created reference architecture (cf. Fig. 1).
The first layer (Access Layer) of the eGov reference architecture concerns access to governmental
services. It addresses the various devices that the end-users utilize. While [16] and [19] also propose
additional channels like the telephone, digital tv, call center, and teleconferencing, our approach is
limited to serving a website through a PC, a mobile device like a smartphone or tablet, or a stationary
kiosk computer that is available, e.g., at a government site. The inclusion of additional mentioned
channels like a call center would add high cost and complexity to the eGov/PB system. The advantage
of the proposed reduction to mobile kiosks and PCs is that setting up a state-of-the-art website that is
also mobile-enabled is sufficient to roll out the application to the targeted devices.
153
Figure 1: Proposed Reference Architecture for a Participatory Budgeting Application in an
eGovernment Systems Landscape
154
The next layer manages the access control through Authentication and Authorization. It builds upon
the single sign-on (SSO) proposed by [13,16] and shall enable the use of all governmental applications
with just one login. It, thus, is a prerequisite for the one-stop paradigm proposed by [14,15,17,19]
(having all necessary resources in one place), enabling proper rights management. There are many
forms one can implement security through authentication. Depending on the implementing
municipality’s legal, technical, and cultural background, one can use login credentials, two-factor
authentication, ID cards, and more. Also, the requirements for participation vary. While some initiatives
might only require a simple registration with a username and password, others might postulate a
residency. After successful authentication, a user is either authorized as a “Citizen” or “Employee” and
can access the parts of the applications that are not publicly available.
The layer Service Bus comprises the business logic of the eGov system. As seen in the literature
analysis, most of the more recent eGov architectures ( [6,8,12,13,15–18]) build upon a Service-oriented
Architecture (SoA)2 to manage the complexity and ensure a high degree of modularity. Almost all other
proposals are not far from the SoA paradigm (cf. interchangeable Java applications [11], portal solution
[19]). Thus, the proposed reference architecture for PB also builds upon the SoA paradigm.
As a result, the PB software in our architecture is just one of many applications that utilize the same
interfaces as the other eGov applications. Our architecture depicts an archetypical PB process based on
[20]. The human pictogram marked with “C” or “P” indicates that authorization is required for the
respective functionalities. In the presented process, only registered “citizens” are allowed to cast a vote,
hand in proposals, and participate in discussions, while registration as a government employee is
necessary to validate and audit the proposals and moderate the discussions.
As already stated, PB initiatives differ widely depending on the jurisdictional and cultural
backgrounds. The instantiation of the application of the PB process is, thus, always bound to local
customization. A detailed description of possible functions along the exemplary process of such a
system is given in [20].
The Data Layer is the gateway for accessing the databases and storing the documents needed for the
applications to fulfill their tasks. The layer is proposed by [7,10,11,13–15]. The data can originate from
systems like an ERP, the central governmental resident register, a PB database, or other application-
specific databases. These databases then build upon the technologies like relational-like databases,
document-centric NoSQL databases, or even Blockchains, e.g., for ensuring secure voting. An audit
trail provides additional security by logging access to security-sensitive data [16].
The Infrastructure Layer is responsible for provisioning the computational resources and
networking infrastructure of the services. The papers upon this reference architecture is built regard this
layer as the provision of servers, networks, data centers, hardware, and more [8,19]. The full or partial
virtualization of these parts of infrastructure through the use of cloud providers can reduce IT costs
while simultaneously improving reliability. A recent literature study on the effects of cloud computing
for eGov outlined its potential [21]. Even though no paper was identified as relevant for eGov
architectures that included a cloud infrastructure, this might be due to the often older literature and the
recent emergence of cloud technologies, combined with the hesitant adaption of new technologies in
governmental branches. As cloud computing and the (at least partially) infrastructure virtualization is
expected to rise in the upcoming years [22], we included this technology in the reference architecture
as a possible deployment model.
The proposed infrastructure layer is based on the NIST definition for cloud computing [23]. It
differentiates four deployment models: In a public cloud, a provider shares computational resources
with the general public, and the servers are on the premises of the cloud providers. In a private cloud,
the virtualization infrastructure is run entirely for (and also possibly by) a single organization. The
community cloud provides a shared infrastructure for exclusive use by selected users (e.g., a
governmental cloud run by the state for administrations). The hybrid cloud combines two or more
infrastructures, e.g., private and public or community and public. This combination allows the execution
of non-sensitive tasks in a public cloud while sensitive data remains within the premises.
2 In a SoA, the PB application is merely one service of many in the government. Even though it can be integrated with other
services, it has a high degree of independence within the system. (As an example, it is not necessarily bound to the same
programming language as the web portal.) It just needs the interfaces to get integrated with the already existing system
landscape.
155
There are three service models regarding the cloud virtualization level: In a Software as a Service
(SaaS) model, the organization orders the whole software from a provider and pays per use (e.g., per
active user). The provider takes care of the provisioning, licensing, and updating of the services. Even
though some applications can also be installed in a private cloud and managed through a provider, these
services normally run off-premise. The provider handles everything except the installed software in a
Platform as a Service (PaaS) model. The managed services include the operating system, runtime
environments, and servers. In an Infrastructure as a Service model (IaaS), the provider handles the
servers, networks, and virtualization infrastructure. Here, the consumer controls everything from the
operating system to the application. However, eGov applications can also run On-Premise without a
cloud virtualization infrastructure. Please note that the decision for or against running the infrastructure
in the cloud is not binary. It is also possible to run just specific, new applications on this virtualization
infrastructure. Most likely, some SaaS application is already in use, like Microsoft Office 365 or Zoom.
4. Conclusion
Managing complex IT architectures is a tedious and complex task, as many systems with many
interfaces need to interact and share data. Without structuring, these systems can uncontrollably grow
out of control, making them costly and hard to maintain. Reference architectures can support the
structuring tasks by providing blueprints. This report presented a reference architecture based on
existing literature for integrating a PB application into an eGov landscape. It is targeted at
administrations who plan to implement a PB initiative. The reference architecture shall enable these
administrations to (1) assess their current eGov architecture, (2) identify the technologies needed to
create a PB, (3) identify possible integrations for a PB application into the existing landscape, and (4)
adapt the proposed reference architecture to their individual needs.
Even though the systematic literature analysis of this paper ensures that the built reference architecture
is based on the already existing body of knowledge, it is not yet validated in practice. Thus, future
research directions should be concerned with validating the proposed reference architecture in real-life
examples.
5. Acknowledgments
The authors are part of the EU-funded EmPaci project (Empowering Participatory Budgeting in the
Baltic Sea Region). This project is part of the European Regional Development Fund.
6. References
[1] A. Shah, Participatory budgeting, World Bank, Washington DC, 2007.
[2] A. Reiz, M. Fellmann, L.-M. Sinervo, A. Heinonen, Selecting the Right Software for Supporting
Participatory Budgeting in Local Government – Reviewing Suitable Solutions, in: Electronic
Governance and Open Society, St. Petersburg, Russia, 2021.
[3] L. Khalid, Software Architecture for Business, Springer International Publishing, Cham, 2020.
[4] S. Angelov, J.J.M. Trienekens, P. Grefen, Towards a Method for the Evaluation of Reference
Architectures: Experiences from a Case, in: R. Morrison, R. Morrison, D. Balasubramaniam, K.
Falkner (Eds.), Software architecture: Second European conference, ECSA 2008, Paphos,
Cyprus, September 29-October 1, 2008 ; proceedings, Springer, Berlin, Heidelberg, 2008,
pp. 225–240.
[5] B. Kitchenham, Procedures for Performing Systematic Literature Reviews. Joint Technical
Report, 2004.
[6] C. Alfaro, J. Gomez, J.M. Lavin, J.J. Molero, A configurable architecture for e-participatory
budgeting support, JeDEM 2 (2010) 39–45. https://doi.org/10.29379/jedem.v2i1.18.
[7] M.Z. Murah, Z. Abdullah, R. Hassan, M. Abu Bakar, I. Mohamed, H. Mohd Amin, Kacang
Cerdik: A Conceptual Design of an Idea Management System, IES 6 (2013).
https://doi.org/10.5539/ies.v6n6p178.
156
[8] BA. Baheer, D. Lamas, S. Sousa, Towards Development of a Reference Architecture for E-
government, in: Proceedings of the 11th International Conference on Theory and Practice of
Electronic Governance, Galway Ireland, ACM, New York, NY, 2018, pp. 640–643.
[9] E. Tambouris, E. Kaliva, M. Liaros, K. Tarabanis, A reference requirements set for public
service provision enterprise architectures, Softw Syst Model 13 (2014) 991–1013.
https://doi.org/10.1007/s10270-012-0303-7.
[10] S.S. Hilabi, F.L. Gaol, T. Matsuo, Enterprise Architecture Design For Regional Parliament
Information Systems Using the Standard Government and Architecture Application (SAGA)
Framework, ICIS Express letters 15 (2021) 178–181. https://doi.org/10.24507/icicel.15.02.173.
[11] R.N. Goddy-Worlu, C.K. Ayo, V.O. Geteloma, A sustainable ubiquitous engagement platform
for open government Implementation, J. Phys.: Conf. Ser. 1378 (2019) 42037.
https://doi.org/10.1088/1742-6596/1378/4/042037.
[12] A. Abdullah, A.-K. Zuhoor, N. Kraiem, Y.A. Jamoussi, Enhanced eGovemment integration
framework for higher interoperability in eGovernment initiatives, in: 2017 International
Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT-
2017), Kannur, IEEE, Piscataway, NJ, 2017, pp. 1322–1329.
[13] M. Al-Husban, C. Adams, Connected Services Delivery Framework, in: Z. Mahmood (Ed.),
Emerging mobile and Web 2.0 technologies for connected e-government, IGI Global, Hershey,
Pennsylvania (701 E. Chocolate Avenue, Hershey, Pa., 17033, USA), 2014, pp. 50–75.
[14] V. Ciobanu, D. Popescu, C. Dobre, F. Pop, V. Cristea, The EU-Services Directive: An E-
framework to Optimize Public Administration, in: 2012 Third International Conference on
Emerging Intelligent Data and Web Technologies (EIDWT 2012), Bucharest, Romania, IEEE,
Piscataway, NJ, 2012, pp. 336–341.
[15] W. Yanyan, Research on E-government Technical Framework Systems, in: 2010 International
Conference on E-Business and E-Government, Guangzhou, China, IEEE, 5/7/2010 - 5/9/2010,
pp. 707–710.
[16] B. Chakravarti, V. Varma, An enterprise architecture framework for building service oriented e-
governance portal, in: 2008 IEEE Region 10 conference, Hyderabad, India, IEEE, Piscataway,
NJ, 2008, pp. 1–6.
[17] F. Corradini, F. de Angelis, C. Ercoli, A. Polzonetti, A Services Management System for Small
and Disadvantaged Communities, in: 2007 29th International Conference Information
Technology Interfaces, Cavtat, Croatia, IEEE Service Center, Piscataway, NJ, 2007, pp. 477–
482.
[18] D. Yang, Y. Han, J. Xiong, eGovGrid: A Service-Grid-Based Framework for E-Government
Interoperability, in: W. Wang (Ed.), Integration and innovation orient to e-society: Seventh IFIP
International Conference on e-Business, e-Services, and e-Society (I3E2007), October 10-12,
Wuhan, China, Springer, New York, NY, 2007, pp. 364–372.
[19] Z. Ebrahim, Z. Irani, E‐government adoption: architecture and barriers, Business Process Mgmt
Journal 11 (2005) 589–611. https://doi.org/10.1108/14637150510619902.
[20] Rostock University, Feature Repository & Recommendations: Group of activities 4.1: Output 3.
Project: “Empowering Participatory Budgeting in the Baltic Sea Region – EmPac”, Rostock,
Germany, 2020.
[21] M. Fathey, F. Olayah, Ali, Abdullah, Gazem, Nadhmi A., The Effect of Cloud Computing
Adoption on the Sustainability of E-Government Services: A Review, International Journal of
Advanced Science and Technology (IJAST) 29 (2020) 2636-2642.
[22] Statista, Worldwide information technology (IT) infrastructure spending breakdown (by value)
from 2014 to 2024, by deployment type, 2021. https://www.statista.com/statistics/486586/it-
infrastructure-spending-forecast-by-type/ (accessed 9 August 2021).
[23] P.M. Mell, T. Grance, The NIST definition of cloud computing, National Institute of Standards
and Technology, Gaithersburg, MD, 2011.
157