A Dissertation About the Job I Was Already Doing

Between 2017 and 2019, while posted to the Fleet Operations division at Northwood Headquarters, I completed a BSc (Hons) in Logistics and Operations Management at Aston University, studying part-time alongside the job. Most of what I have written on this site covers 2023 onward: the MIT MicroMasters, the Defence Logistics Staff Course, the case studies drawn from operational assurance work. The degree stands behind all of it.

Three Years, Alongside the Job

The degree was delivered through Aston University's School of Engineering and Applied Science, taken part-time over three years while I remained in post at Northwood. Six taught modules ran across the first two years, followed by a final year dissertation module, EC310A, and the degree was awarded at 2:1.

The degree was worth doing properly because the coursework overlapped so directly with the work in front of me. The study took place inside a Fleet Operations department already responsible for the kind of planning the modules described, so each assignment could be tested against live practice as it was written.

Six Modules, Building Towards the Same Argument

Each module produced a substantial written assignment, and looking back at them together shows a consistent thread running through what might otherwise read as six unrelated pieces of coursework.

Multi-Modal Transport Management examined a proposed modal shift of container transport within the UK, comparing unimodal and multimodal networks and the case for intermodal solutions. Inventory Management used a manufacturing case study to work through operations strategy, line sequencing, and the comparative value of vendor-managed versus customer-managed inventory. Sustainability in Logistics Activities looked at reverse logistics and closed-loop supply chains in a military context, the process the armed forces call retrograde: recovering, repairing, and redistributing assets rather than replacing them.

Lean, Agile and Competitiveness critiqued lean and agile approaches to supply chain management across different business sectors. Leadership and Human Resources Management, taken in the final year, examined organisational resistance to change through Lewin's and Kotter's models. Financing the Supply Chain, also final year, covered the Defence Acquisition Trade Cycle: strategic procurement, Smart Acquisition and Through-Life Support, Contracting for Availability, integrated logistics support, and how each is measured.

Two of those six modules turned out to be a first draft of arguments I would return to years later. The reverse logistics assignment is, in substance, the same question I examined again in 2025 on the Defence Logistics Staff Course, comparing the MoD's reverse supply chain against automotive sector practice. The Defence Acquisition Trade Cycle assignment covers much of the same ground as the Financing Defence Logistics and Procurement Management modules I am studying now. The same questions have now run through seven years of study.

The Dissertation: Forecasting Fuel for a Fleet

The final year project examined the forecasting of fuel consumption for Royal Navy ships, supervised by David J. Carpenter. At the time, National Audit Office reporting put fuel as a significant share of the RN's annual operating budget, in the tens of millions of pounds (National Audit Office, 2011). Action Research was chosen as the methodology, following the framework Saunders, Lewis and Thornhill describe as the Research Onion (Saunders, Lewis and Thornhill, 2016). I held a direct operational role in the department the research examined, and my own line management had proposed the topic as an area where they wanted innovative thinking.

Primary data came from semi-structured, open-ended questionnaires with senior stakeholders directly responsible for fuel forecasting and data management, gathered under a formal informed-consent process through Aston's ethics guidelines. A Commander at the Maritime Operations Centre, Fleet Operations division at Northwood, a researcher and the staff of a Maritime Warfare Centre, and a Lieutenant Commander all gave their time and expertise, for which I remain grateful. Additional conversations with engineers working on the Navy's Project Nelson initiative, then in its early stages, shaped the technical direction of the recommendation directly. Secondary research drew on National Audit Office reports into the MoD's supply chain and financial management (National Audit Office, 2011), and the Ministry of Defence's Joint Concept Note on human-machine teaming (Ministry of Defence, 2018).

The findings were consistent with what the National Audit Office had already found at organisational level. Fuel forecasting depended on a manual reporting process with no electronic or real-time data source and no central repository. Strategic budgeting, operational planning and procurement teams each worked from separately maintained copies of the same underlying figures. Forecast error translated directly into cost, either as in-year underspend that could have been allocated elsewhere, or as the need to source additional fuel at short notice and at a price premium.

The recommendation was a single, integrated, service-oriented data architecture to replace the patchwork of legacy systems. It was built around using artificial neural networks against the fleet's fuel-consumption data, the same machine learning direction the Navy was exploring at the time through Project Nelson. Read now, it is an early and fuel-specific version of an argument that runs through this site. Forecasts, decisions and assurance are only as reliable as the data feeding them, and the data problem is often the problem underneath the process problem.

What Carried Forward

This degree is the piece that was missing from the published account of 2017 to 2019. It was completed part-time and examined against a real operational problem before Aston examined it. It rests on the same argument about data and decision-making that runs through everything I have written since. The Defence Logistics Staff Course gave that argument a second, more senior audience, years after a dissertation had already made the case.

The neural-network proposal also started my path into machine learning, prompted by a fuel-forecasting problem that needed it. Jovian's deep learning course followed the degree, then a MicroMasters in predictive analytics with the University of Edinburgh, both continuing the direction this dissertation first pointed towards. The degree is the starting point of the argument about data and decision-making that the later work on this site develops.