Forecasting digital human capital requirements for Ghana’s energy transition: A machine learning and structural equation modeling approach

Title : Forecasting digital human capital requirements for Ghana’s energy transition: A machine learning and structural equation modeling approach

Abstract:

Ghana's National Energy Transition Framework (2022–2070) sets out an ambitious pathway toward decarbonisation through investments in renewable energy, grid modernisation, electric mobility, and clean cooking. Delivering this transition will require not only infrastructure and finance but also a digitally skilled workforce capable of operating an increasingly digitalised and data-driven energy system. At present, however, the digital skills dimension of Ghana’s energy transition has not been sufficiently quantified in national manpower planning. This is a concern given high youth unemployment and rising demand for digital skills across millions of emerging jobs by 2030. The result is a likely structural mismatch between required and supplied skills in the domestic education and training system. Existing studies on energy transition employment are often aggregate and technology-focused, while studies on digital human capital are usually sectorneutral and rarely linked to workforce forecasting. This study develops an integrated machine learning–structural equation modeling (ML–SEM) approach to forecast Ghana's digital human capital requirements for the energy sector through 2030. The machine learning component will use historical and cross-sectional labour, education, and energy-sector data to forecast digital skills demand by occupational category. The structural equation modeling component examines institutional, organisational, and educational factors shaping digital skills adoption in energysector employers and training institutions. The two components are integrated within a sequential explanatory design in which machine learning forecasts quantify future digital skills demand, while the structural model identifies the institutional and educational factors that influence the sector's capacity to meet that demand. 

The study will provide a sector-specific forecast to inform curriculum reform, technical and vocational education, and industry–training partnerships. This study is among the first to integrate machine learning-based labour demand forecasting with structural equation modeling to support digital workforce planning for Ghana's energy transition.

Biography:

He is an experienced finance professional and Chartered Accountant with over 10 years of postqualification experience, a Fellow of ACCA, and a member of the Institute of Chartered Accountants, Ghana. He hold an MBA in Financial Management from London Metropolitan University and an MA in Economic Policy Management from the University of Ghana, and He is currently a PhD candidate in Energy Economics. His professional experience spans the oil and gas, mining, engineering, manufacturing, banking, and FMCG sectors. His research interests include fossil fuel taxation, energy market shocks, energy transition investment, risk and return analysis, machine learning applications, and energy demand and supply forecasting.