Solar photovoltaic installations have now become a common sight across the globe. However, in places with a high level of dust, the panels have not performed as expected. The dust deposition acts to reduce the effective light that the solar cells receive thus reducing the output. As such, cleaning regimes are instituted to improve the performance of the panels. The more often the cleaning, the better the performance, but this could likely imply more energy is used for cleaning and the improvement may not be sufficient. In this paper, we develop a model to forecast the output power with respect to variables of cleaning, wind, humidity and temperature. The result shows that it is possible to develop a model to optimise cleaning regimes taking into account meteorological factors. One has to consider the impact of wind, which was found to be positively correlated with photovoltaic power output.