Designing a time-driven ABC model: Reducing the number of time equations though business analytics

Abstract

Purpose The purpose of this paper is to investigate how business analytics, statistical learning and multiple regression analysis can streamline time equations within the Time-Driven Activity-Based Costing (TD-ABC) model. This study contributes to the TD-ABC literature by identifying statistical relationships between resource usage, time equations and costs. Design/methodology/approach This study uses a range of statistical techniques, culminating in multiple regression analysis, to determine which time equations are most relevant for decision-making within a TD-ABC framework. Findings Using a combination of synthetic and experimental data, the author demonstrate how statistical learning and regression analysis can identify the most impactful time variables for decision-making. This approach enables a perceptible reduction in the number of time equations by focusing only on those variables that are statistically significant and relevant for future decisions. Research limitations/implications As this study is based on a hypothetical TD-ABC layout and a data management approach, future research could explore the application of this statistical model in collaboration with real-world companies. Practical implications Limiting the number of time equations through regression analysis can reduce the number of time equations, complexity and cost of implementing TD-ABC. Moreover, the statistical approach supports a “cause-and-effect” assumption that underpins the TD-ABC methodology. Originality/value Beyond introducing a statistical approach to time equation modeling, this paper documents and discusses the process of identifying relevant time variables in a TD-ABC environment, offering a novel perspective on cost modeling and decision support.