Nowadays, with mechanization development, evaluation of machine productivity and calculation of average cost of skidding for each hour and each cubic meter is necessary to compare their efficiency. In this study, the mathematical model of production and cost of steel tracked skidder LTT-100A was studied in Tarbiat Modares University Experimental Forest (TMUF). Continuous time study was applied during skidding operation to record the time spent for each component of skidding. The obtained data was then analyzed using multivariable regression analysis to build the mathematical model of production. Results showed that skidding distance, skid trail slope and volume of logs at each cycle are significant variables influencing skidding time. The share of delay was 14.94 % of total time while 44% of delays time was identified as technical delay (unavoidable delay), personal and operational delay were recorded 33% and 21% percent, respectively. Skidder productivity regardless of delay time and effective time were calculated 5.84 m3/h and 4.98 m3/h, respectively. Total average cost of system was estimated 265678 RLS/h and 54330 RLS/m3, respectively (Rials= 50cent).