Optimizing operational efficiency and measurement integrity in Uganda’s fuel tanker logistics.
| dc.contributor.author | Nsimbe, John | |
| dc.date.accessioned | 2026-10-08T11:49:50Z | |
| dc.date.available | 2026-10-08T11:49:50Z | |
| dc.date.issued | 2026-09 | |
| dc.description | Dissertation | |
| dc.description.abstract | This study develops a data-driven, mixed-methods decision-support framework that integrates operational variables with metrological variables, statistical analysis, qualitative coding (NVivo 15) and a Genetic Algorithm (GA) implemented in Python 3.12 and developed in VS Code. Field data (dispatch & delivery volumes, temperatures, densities, seal status, delays, telemetry uptime, safety scores, and stakeholder feedback) was collected from 106 trucks from Busia OSBP, Kampala depots, and fuel stations. All volumes were corrected to a 20 °C reference using fuel-specific thermal expansivity and density adjustments. Results indicated average fuel loss in the range of (1%–5%), with site patterns showing Busia OSBP (average absolute loss approx. 802.1 L), Kampala depots (approx. 500 L), and fuel stations (approx. 112.3 L). NVivo 15 coding corroborated qualitative priorities where Fraud & Siphoning Risks and Calibration & Measurement Integrity were the most frequently coded themes across depot managers, drivers, OMC staff, and end customers. Statistical analysis produced compressed component scores which were used as objective weights in the GA fitness function that balanced operational efficiency and measurement integrity. The GA converged on an optimized loss mitigation package termed as best solutions where PCA-derived component scores (FAC1 = - 2.16074, FAC = 2.000, FAC 3 = 0.96198) were used as objective weights in the fitness function balancing efficiency and integrity giving a best Loss = 43.0476 L and Best Fitness = 0.20897. These prioritized calibration scheduling for high-risk tankers and custody transfer points, automated volume correction factors and telemetry reconciliation at handovers and tamper-evident digital seals integrated with telemetry reconciliation rules. This combined strategy was selected because it simultaneously reduces meter drift exposure, enforces corrected volumetric accounting, and closes primary siphoning pathways. Scenario simulations and sensitivity tests show consistent improvements in delivery accuracy and reductions in expected loss compared with baseline heuristics. The final deliverable is a reproducible Python module and a ranked set of implementable interventions that together improve measurement integrity, reduce fuel losses, and strengthen regulatory confidence in Uganda’s downstream fuel tanker logistics | |
| dc.description.sponsorship | Dr. Peter Tumutegyereize ; Dr. Joseph Ddumba Lwanyaga ; Busitema University | |
| dc.identifier.citation | Nsimbe, J. (2026). Optimizing operational efficiency and measurement integrity in Uganda’s fuel tanker logistics. [Unpublished dissertation]. Busitema University. | |
| dc.identifier.uri | https://bdears.busitema.ac.ug/handle/123456789/10531 | |
| dc.language.iso | en | |
| dc.publisher | Busitema University | |
| dc.title | Optimizing operational efficiency and measurement integrity in Uganda’s fuel tanker logistics. | |
| dc.type | Other |