A COMPARATIVE ANALYSIS OF PREDICTIVE MODELS FOR HOUSEHOLD ENERGY CONSUMPTION IN LAGOS

• Raheem, W. A.,

  University of Lagos, Lagos State, Nigeria

• Omiyale, A. D

  University of Lagos, Lagos State, Nigeria

• Odeyinka, O. F.

  Corona College of Education, Lagos, Lagos State, Nigeria

• Folorunso, C

  Corona College of Education, Lagos, Lagos State, Nigeria

Rapid urbanisation, unreliable electricity supply, and increasing reliance on electrical appliances have heightened residential energy demand and created the need for accurate prediction of household energy usage. While deep learning and hybrid predictive models have been increasingly applied to energy forecasting globally, their use in African contexts remains comparatively limited, with relatively few studies addressing the added complexities of infrastructural fragility and diverse user behaviours. In Nigeria, there has been limited effort in comparative modelling research, particularly at the household level, where energy behaviour is shaped by a complex interplay of socio-economic and infrastructural factors. This study examines household energy consumption patterns in Lagos, Nigeria, using a comparative, multi-model approach. Data from 350 households, including socio-economic status, appliance usage, billing methods, and conservation practices were collected and analyzed. The results of four predictive models, Linear Regression, Artificial Neural Networks (ANN), AutoRegressive Integrated Moving Average (ARIMA), and a hybrid ARIMA–Long Short-Term Memory (LSTM) model, were compared. Findings indicate that household income, dwelling type, and appliance usage are significant predictors of electricity consumption. Among the models, the hybrid ARIMA–LSTM achieved the best performance (RMSE of 580, MAE of 400, R² of 0.85), combining the strengths of both linear and non-linear modeling. The ANN model also demonstrated strong predictive accuracy (R² of 0.79), while linear regression was limited by its inability to capture complex relationships. The results highlight the potential of integrating machine learning into urban energy planning. Policymakers could use these insights to support targeted interventions such as smart metering, appliance subsidies, and behaviour-based demand management.

1. Raheem, W. A., , University of Lagos, Lagos State, Nigeria

           Department of Systems Engineering, University of Lagos, Lagos State, Nigeria

2. Omiyale, A. D, University of Lagos, Lagos State, Nigeria

   Department of Systems Engineering

3. Odeyinka, O. F. , Corona College of Education, Lagos, Lagos State, Nigeria

   Department of Computer Science Education

4. Folorunso, C, Corona College of Education, Lagos, Lagos State, Nigeria

   Department of Computer Science Education

• pdf

Copyright (c) 2025 FUTA JOURNAL OF ENGINEERING AND ENGINEERING TECHNOLOGY

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Copyright

With the submission of a manuscript, the corresponding author confirms that the manuscript is not under consideration by another journal. With the acceptance of a manuscript, the Journal reserves the exclusive right of publication and dissemination of the information contained in the article. The veracity of the paper and all the claims therein is solely the opinion of the authors not the journal.