How User Expectations for Features and Services Shape Computing System Design in Software-Defined Vehicles: Implications for Energy Consumption and Environmental Sustainability

Abstract

The automotive industry is undergoing a profound transformation with the advent of Software-Defined Vehicles (SDVs), driven by escalating user expectations for advanced features, personalized digital experiences, and enhanced connectivity. This dissertation investigates how these evolving user demands fundamentally shape computing system design in SDVs and critically evaluates the resulting implications for vehicle energy consumption and environmental sustainability from a strategic business perspective.
To address this, a practitioner-led, mixed-methods approach was employed. Quantitative analysis was performed on extensive vehicle telemetry and power consumption data, allowing for precise measurement of energy footprints across different SDV configurations. This was complemented by qualitative insights gathered through structured interviews with industry experts and in-depth customer studies, providing a nuanced understanding of user expectations and market dynamics.
Environmental impacts were quantified using a comprehensive Life Cycle Assessment (LCA) framework, specifically focusing on the operational phase and evaluating the influence of country-specific electricity mixes on emissions. To facilitate this analysis, SDVs were systematically categorized into Entry-, Mid-, and Premium-Levels, enabling a detailed examination of their energy footprint across various System-on Chip (SoC) architectures (Single, Dual, Multi-SoC).
Key findings reveal a significant, non-linear increase in computational demands and power consumption as SDV sophistication rises, with Premium-Level SDVs and multi-SoC architectures exhibiting substantially higher energy requirements and associated CO2 emissions. Crucially, the embodied carbon footprint of an SDV is not intrinsic to the technology itself but is profoundly influenced by the energy mix of its manufacturing location and the interplay of socio-economic factors and policy landscapes. The analysis and discussion around it also highlight a critical paradox; while clean energy grids enable low-emission production, economic barriers can limit the adoption of advanced SDVs in certain regions, and high-income markets may prioritize feature richness over sustainability in the absence of robust policies.
To summarize, this thesis contributes an integrated theoretical framework linking user expectations, architectural decisions, and environmental outcomes, extending literature on demand-driven innovation and sustainable digital transformation. Managerially, it provides actionable guidance for SDV manufacturers on portfolio prioritization, compute-resource governance, supplier collaboration, and emissions-reduction strategies that balance user-centric innovation with critical ESG commitments, informing both corporate strategy and industry-level standards for responsible SDV development.
Keywords: Software-Defined Vehicles, User Expectations, Computing System Design, Energy Consumption, Environmental Sustainability, Life Cycle Assessment, Automotive Strategy, Embodied Carbon, SoC Architecture.