Journal Design Engineering Masthead
African Civil Engineering Journal | 08 December 2006

A Time-Series Forecasting Model for Yield Improvement in Nigeria's Industrial Machinery Fleets

A Policy Analysis for Strategic Maintenance Optimisation
C, h, i, n, w, e, i, k, e, O, k, o, n, k, w, o, ,, A, m, i, n, a, S, u, l, e, i, m, a, n, -, B, e, l, l, o
Predictive MaintenanceSARIMA ModellingIndustrial PolicyAsset Management
SARIMA model provides statistically significant predictive capability for machinery failure windows.
Shift from reactive to predictive maintenance enables evidence-based policy formulation.
Yield improvement contingent on systematic data collection and institutional adoption.
Model offers quantitative basis for moving beyond schedule-based maintenance regimes.

Abstract

{ "background": "Persistent low yield from industrial machinery fleets in Nigeria's manufacturing and construction sectors represents a critical constraint on economic development. Current maintenance policies are largely reactive, leading to excessive downtime and capital inefficiency. A shift towards predictive, data-driven policy is required.", "purpose and objectives": "This policy analysis evaluates a novel time-series forecasting model designed to measure and improve machinery yield. The objective is to provide a methodological framework for strategic maintenance optimisation, enabling evidence-based policy formulation for fleet management.", "methodology": "The analysis employs a Seasonal Autoregressive Integrated Moving Average (SARIMA) model, formalised as $\\phi(B)\\Phi(B^s)\\nabla^d\\nabla^Ds Yt = \\theta(B)\\Theta(B^s)\\epsilon_t$, applied to historical operational availability and output data from a representative sample of machinery fleets. Model diagnostics include analysis of robust standard errors to assess parameter stability.", "findings": "The forecasting model demonstrates a statistically significant predictive capability for machinery failure windows, with a lead time sufficient for proactive intervention. Application of the model to simulated policy scenarios indicates a potential yield improvement of 18-24% through optimised maintenance scheduling, contingent on data quality and institutional adoption.", "conclusion": "The integration of time-series forecasting into maintenance policy presents a viable pathway for substantial yield gains. The model provides a quantitative basis for moving beyond schedule-based maintenance regimes, though its efficacy is dependent on systematic data collection and workforce upskilling.", "recommendations": "Policymakers should mandate the standardised collection of machinery performance data. A pilot programme for model implementation in state-owned enterprises is advised. Investment in training for predictive maintenance analytics is essential for long-term sustainability.", "key words": "Predictive maintenance, SARIMA modelling, industrial policy, asset management, operational research", "contribution statement": "This paper provides the first applied framework integrating SARIMA forecasting directly into national industrial maintenance policy for Nigeria, demonstrating a concrete methodology to translate operational data into