Abstarct: Open-pit mining with high production rate requires large machineries. Deployment and maintenance of these machineries require huge investment. Unit operations of mining involve drilling, blasting, loading and hauling which loading occupies a crucial role in production rate and if it stops, the whole operation will stop. Unforeseen stops will increase maintenance and repair expenses, decrease profit, delay fulfilling the orders, and ultimately reduce the company's credit and will lead to customer dissatisfaction. Spare parts management is part of product support and is very effective in product's life cycle cost. Availability of spare parts could decrease the system downtime and increase its efficiency which result in improvement of project profitability. In this thesis, the reliability index is used for unloading spare parts. On completion of this approach, the impact of environmental conditions added to this analysis as "observed risk factors" to make the results as close as possible to reality. If we miss considering the harsh working conditions in mines, there will be a major defect in our analysis. however we know that it is impossible to determine and import all the observed risk factors in the analysis because some risk factors such as managerial decision-making, difference of brands, or the skill of operators, despite being observable, have not been recorded or correctly categorized and remained unknown due to various reasons including impossibility of quantification, lack of both information and accuracy, lack of expertise and shortage of time in data collection interval. These risk factors also added to the analysis as "unobserved risk factors". To give more consideration to the effect of observed and unobserved risk factors on reliability function (estimation of related parameters) and ultimately the spare parts provision baxsed on resultant reliability function, we suggest a five-step algorithm. For this purpose the expansions baxsed on Cox regression models have been used, because these models are able to simultaneously analyze the performance data (time data) and effective environmental conditions (observed and unobserved risk factors). Thereafter, in order to validate the suggested algorithm for spare parts provision baxsed on observed and unobserved risk factors, a case study of the loading fleet (Caterpillar 390DL bucket nail excavator, Golgohar Sirjan Iron Mine No. 1) was selected. After establishing the databaxse within 24 months for 4 excavators, the required data collected in two formats: time data including time to failures (TTFs), and observed risk factors including working shift (Z1), rock type (Z2), operation team (Z3), temperature (Z4), precipitation (Z5) and Excavator code (Z6). The value of maximum likelihood ratio equals 23.56, which confirm the existence of unobserved risk factors. Therefore, the exponential mixed proportional hazard model (MPHM) with 0.297 as the frailty value used for the reliability function. The results of spare parts management with the mixed proportional hazard model in one year for the existing scenarios are 76, 129 and 222 respectively, and the classical exponential function predicted to be 112. Finding of this study shows the significant effect of unobserved risk factors on spare parts analysi.