2. Professional Disclosure
2.1 Author & Organisational Background
My name is Akeju, Oluwafemi Olusoji, and I serve as the Chief Operating Officer (COO) of KK Leasing Limited, a Nigerian third-party logistics (3PL) and vehicle leasing company. In this capacity, I bear executive responsibility for the end-to-end management of the company’s vehicle fleet — which constitutes its primary revenue-generating and cost-bearing asset class. My remit encompasses fleet procurement, maintenance governance, workshop operations, client SLA management, fuel and tyre budgeting, and the strategic deployment of vehicles across multiple corporate client accounts.
KK Leasing’s core value proposition is to provide, maintain, and manage vehicles for corporate organisations, thereby enabling our clients to outsource all fleet-related operational risk and complexity to us. The company’s financial performance is, in large measure, a direct function of how efficiently we control maintenance costs relative to the revenue generated from each vehicle. A vehicle grounded for unplanned repair neither generates revenue nor meets client SLA commitments — making maintenance analytics not merely an academic exercise but an existential operational priority.
2.2 Why This Analysis Matters Operationally
The maintenance function at KK Leasing is characterised by high frequency, significant value, and considerable complexity. In 2023 alone, the dataset reflects over ₦93 million in approved maintenance expenditure across 153 vehicles, 22 distinct maintenance categories, and 6 client accounts. Without structured analytical oversight, cost overruns go undetected, high-risk vehicles remain in service beyond economic useful life, and budget forecasts are based on intuition rather than evidence.
This analysis directly supports three strategic decisions I confront on a recurring basis:
- Vehicle retirement decisions — When does a vehicle’s cumulative maintenance cost exceed its residual value and warranted budget allocation?
- Maintenance budget allocation — How much should be budgeted per vehicle per category, and which categories carry the highest financial risk?
- Workshop sourcing strategy — Should maintenance be directed to in-house workshops or third-party vendors for optimal cost and quality outcomes?
2.3 Technique Justification
The five analytical techniques selected for this study are not merely textbook exercises — each maps directly to a recurring operational question in the management of KK Leasing’s fleet.
Technique 1 — Exploratory Data Analysis (EDA) Before any inference or prediction, a rigorous analyst must first understand the data they possess. In the fleet management context, EDA is indispensable for identifying the baseline cost landscape: which maintenance categories consume the largest share of the budget, how costs are distributed across vehicles and months, and where data quality issues (such as missing values, duplicate entries, or outlier repair events) may distort downstream analysis. Our maintenance cost data is inherently right-skewed — a small number of high-value repair events (e.g., engine overhauls, tyre fleet replacements) dominate total spend. EDA allows me to detect and document these anomalies before they corrupt statistical conclusions. It also generates the summary statistics — means, medians, standard deviations, and inter-quartile ranges — that form the factual basis for budget conversations with the Board and with clients.
Technique 2 — Data Visualisation Numbers alone do not persuade operational managers or clients. Visualisation transforms raw maintenance data into narratives that are immediately actionable. In my role, I regularly present fleet performance dashboards to the management team and to client procurement committees. The grammar-of-graphics approach adopted here — using deliberate chart type selection, consistent colour coding, and layered storytelling — allows a non-technical audience to grasp, for instance, that tyre expenditure is the single largest cost category, that costs peak in the mid-year months, or that certain vehicle makes exhibit disproportionate maintenance intensity. These visual insights directly inform decisions on preventive maintenance scheduling, tyre procurement contracts, and vehicle replacement cycles.
Technique 3 — Hypothesis Testing Operational intuition frequently generates claims that need rigorous verification: “Toyota vehicles cost more to maintain than other makes”, or “mechanical repairs are significantly more expensive than electrical ones.” Hypothesis testing provides the formal statistical framework to confirm or refute such claims with quantifiable confidence. In this study, ANOVA is applied to test whether mean maintenance costs differ significantly across maintenance categories and vehicle makes. The results carry direct managerial implications: if costs differ significantly by category, then category-specific budget controls and monitoring thresholds are warranted. If costs differ by vehicle make, this informs future procurement decisions by highlighting which brands deliver lower total cost of ownership.
Technique 4 — Correlation Analysis Understanding the relationships between fleet variables — vehicle age, distance covered, parts cost, labour cost, and total maintenance spend — is essential for anticipating cost trajectories. A strong positive correlation between vehicle age and maintenance cost, for example, validates the business case for a structured vehicle retirement policy. Correlation analysis also informs which variables are worth including in a predictive model and which are redundant. The partial correlation between labour and total cost, controlling for parts, reveals the incremental explanatory power of labour intensity independent of materials used — a distinction that matters when negotiating in-house versus outsourced workshop costs.
Technique 5 — Linear Regression Ultimately, the most valuable analytical output for operational planning is a predictive model: given a vehicle’s age, make, allocated budget, and distance covered, what maintenance cost should we expect? Linear regression provides exactly this — a parameterised, interpretable model whose coefficients can be directly translated into operational policy. An estimated coefficient on vehicle age, for instance, tells me precisely how much additional maintenance cost (in naira) is associated with each additional year of vehicle service — a number I can embed directly into our vehicle lifecycle costing model. Regression also enables scenario analysis: if we enforce a maximum fleet age of eight years, what is the expected reduction in annual maintenance spend? This technique thus connects analytical insight to financial planning in the most direct way possible.
References
Adi, B. (2026). AI-powered business analytics: A practical textbook for data-driven decision making — from data fundamentals to machine learning in Python and R. Lagos Business School / markanalytics.online. https://markanalytics.online
Akeju, O. O. (2026). KK Leasing Limited fleet maintenance transaction records, January–November 2023 [Dataset]. Operations Department, KK Leasing Limited, Lagos, Nigeria. Data available on request from the author.
KK Leasing Limited. (2023). Vehicle fleet master file and maintenance budget extract, FY2023 [Internal data]. Operations Department, KK Leasing Limited, Lagos, Nigeria.
R Core Team. (2024). R: A language and environment for statistical computing (Version 4.4). R Foundation for Statistical Computing. https://www.R-project.org/
Van Rossum, G., & Drake, F. L. (2009). Python 3 reference manual. CreateSpace.
Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L., François, R., Grolemund, G., Hayes, A., Henry, L., Hester, J., Kuhn, M., Pedersen, T. L., Miller, E., Bache, S. M., Müller, K., Ooms, J., Robinson, D., Seidel, D. P., Spinu, V., … Yutani, H. (2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686. https://doi.org/10.21105/joss.01686
Wickham, H. (2016). ggplot2: Elegant graphics for data analysis. Springer. https://doi.org/10.1007/978-3-319-24277-4
Allaire, J. J., Teague, C., Scheidegger, C., Xie, Y., & Dervieux, C. (2022). Quarto (Version 1.x) [Computer software]. https://doi.org/10.5281/zenodo.5960048
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, É. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.
McKinney, W. (2010). Data structures for statistical computing in Python. In Proceedings of the 9th Python in Science Conference (pp. 56–61). https://doi.org/10.25080/Majora-92bf1922-00a
