MASTER Assignment
An ENTERPRISE ARCHITECTURE based Big Data Analytics Deployment Reference Architecture to improve Business Value
Type : Master M-BIT
Period: Mar, 2023 - Aug, 2023
Student : Contecha Montes, J.A. (Jesús, Student M-BIT)
Date Final project: August 14, 2023
Supervisors:
Abstract:
The increasing architectural complexity poses a significant challenge for digital leaders as organizations are at risk of being overwhelmed by data floods, complexity, and rising costs. As companies transition to become AI-driven entities, the architectural complexity and IT fragmentation increase. Furthermore, while structured data storage is projected to increase, a significant portion of data stored remains unused. Moreover, only half of Data & Analytics teams effectively contribute value to their organizations, suggesting that a fraction of available structured data is used to create incremental business value, indicating data and resources underutilization. To overcome these challenges, Enterprise Architecture (EA) artifacts are proposed to serve as a "blue print roadmap representa- tion" to guide the deployment of Big Data analytics (BDA) initiatives. However, further research is needed to understand the role of EA in adopting big data analytics. This study aims to integrate the Big Data Analytics capabilities theory, EA frameworks, and empirical organizational resources by exploring how Enterprise Architec- ture can improve the deployment of big data analytics initiatives. EA plays is theorized to play a critical role in representing digital transformation’s building blocks and pro- cesses to align Information Systems with business strategy. This multifaceted approach implies a reference architecture that captures business, applications, and information and technology architectures changes. The present thesis emphasizes the importance of EA practice in planning, guiding, and assessing the transformations required to leverage current and future BD capabilities and resources to develop DB Analytics capability. By effectively leveraging EA artifacts, orga- nizations can orchestrate big data resources (People, Process, Tech) and BDA capabilities (Business Infrastructure alignment, seizing/reconfiguration, and Infrastructure Flexibil- ity) to optimize deployment processes. For instance, a firm/function where "data validates business experience," versus "business experience complements data" would benefit from a specific deployment architecture tailored to its capabilities and resources context. To this end, three architectural levels are developed to represent essential deployment processes and core building blocks that serve as a blueprint for big data initiatives. Finally, concrete architecture levels are instantiated and evaluated within Heineken’s Advanced Analytical product global deployment context. The results indicate that the architecture effectively encompasses core components, enables cross-functional teams, and reduces deployment time and improve resource optimization.