Enhancing fuel consumption efficiency is an essential topic for operational sustainability and operational cost in the mining industry. In open pit mining, one of the biggest contributors to fuel consumption is dump trucks (DT). The varying operator driving behaviour and frequently changing mine road environmental conditions, such as road surface and road grade, lead to often unmanageable DT operations and result in sub-optimal fuel consumption. Therefore, the author aims to develop a guidance system that can manage the driving behaviour of DT operators. This study proposes a guidance system that uses a dynamic analytical model with customised algorithms and a near real-time telemetry system to adapt to the changing mine road environment. Using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm combined with other data processing techniques, the model can determine fuel-efficient driving behaviour represented by speed, engine speed, and accelerator position and can adapt to changing mine road environmental conditions without compromising DT productivity. In addition, the model runs every three hrs to increase the model's flexibility to changes in the mine road environment. The authors evaluate the model using two parameters, the cycle fuel (L) which is a parameter that calculates fuel consumption in a mining road segment and the normalised productivity metric (m3/h.km). In other words, the cycle fuel indicates the actual fuel consumption, and the normalised productivity metric indicates the efficiency of the fuel consumption. From the evaluation results, this approach was able to produce an average fuel saving of 2.42 L/h per unit or 75 015.1 L in total. In addition, the higher adherence group has a higher productivity by 11.6 per cent compared to lower adherence groups. This result explains that the model developed by the authors, not only produces fuel saving, but also enhances the fuel consumption efficiency of DT at open pit coalmine sites. This paper details the algorithm specification and model development, deployment of the model, and the closed loop workflow of the model.