Abstract: Maintenance cost of agricultural machines accounts for about 10 to 15% of total production cost and this tends to increase as the machine gets older. To stabilized this already high cost, agricultural machinery maintenance engineers chose Preventive Maintenance (PM) strategy. However, the latter may not produce the desired results especially when machines are used under different operating context. In this paper, an assessment of the effects of PM strategy on reliability indicators (failure rate, mean time between failures (MTBF), mean time to repair (MTTR) and mechanical availability) of a rice combine harvester Mitsubishi TVR96 working under the operating context of Pacharr rice fields in The Gambia, is discussed. These reliability indicators have been computed for each month of the harvesting seasons of 3 years; 2013, 2014 and 2015. The failure rate of the machine reaches an all-time high in December of 2015 with 0.889, MTBF was between 1.13 hour and 17 hours, MTTR was between 0.86 hour and 5.63 hours and the mechanical availability fluctuate with the highest (94%) recorded in June of 2013 and the lowest (44%) recorded in December of 2015. These unfavourable results prompted us to propose a new PM strategy, condition based maintenance (CBM). The results of this proposed strategy show an improvement of the various reliability indicators; failure rate dropped between 0.014 and 0.033, MTBF increases to between 30.63 hours and 70.02 hours, MTTR drops to between 0.62 hour and 0.79 hour and the mechanical availability increases to between 98% and 99%. This strategy will be implemented on the machine in 2018, 2019 and 2020 to see if same results will be obtained.