A reference architecture for implementation of big data analytics by public sector organizations in developing countries: a case study of Uganda

Abstract

Despite the growing popularity of Big Data Analytics (BDA) technologies, many BDA implementation projects in public sector organizations fail or fall short of expectations. Moreover, several public sector organizations are uncertain of how to approach BDA implementation. The implementations of BDA technologies have presented many opportunities to public sector organizations of all sizes all over the world. BDA technologies like Hadoop Distributed File systems and Cassandra have unlocked several opportunities for better-quality public service delivery, informed policy-making, and citizenry participation among others. BDA technologies allow for effective strategic decision-making by uncovering insights and patterns. It is stated that Big Data (BD) can only benefit society if there are effective tools that help organizations to make sense of the increasingly voluminous data. While public sector organizations in developed economies have incorporated BDA implementation to advance public service delivery, the reverse is true for developing economies. The failure rate of BDA implementation projects is thought to be over 75%, and many people are unsure of how to approach them. Among the challenges, reference architectures are frequently mentioned. Public sector organizations can greatly improve their implementation success of BDA implementation if they incorporate the BDA reference architectures into their BDA implementation initiatives. The general objective of the study was to develop a reference architecture for the implementation of BDA by public sector organizations in developing countries taking a case study of Uganda. The study adopted the Design Science Research Methodology (DSRM) because the research aimed at developing an artifact in the form of reference architecture for the implementation of BDA by public sector organizations in developing countries. A survey was conducted using a questionnaire among nineteen (19) respondents from 3 case public sector organizations in Uganda to understand the BDA implementation issues in public sector organizations. Survey findings revealed the requirements needed for the implementation of BDA in public sector organizations in developing countries from which design decisions were derived and used to design the reference architecture for implementing BDA by public sector organizations in developing countries, taking Uganda as a case study. The designed reference architecture improves and builds on NIST reference architecture. The reference architecture was evaluated using Structured Walkthroughs in 1 phase and was done with 20 IT experts. The evaluation results indicated that the reference architecture was functional, understandable, and traceable for guiding public sector organizations in developing countries to implement BDA as a way to utilize BD.