OPTIMISATION OF SHOVEL-TRUCK HAULAGE SYSTEM IN AN OPEN-PIT USING QUEUING APPROACH

Abstract

Equipment selection is key activity in mining operation because it accounts to more than 60% of total operation cost. When selection of equipment is not properly done it results into over- trucking or under-trucking. Under-trucking reduces loader utilisation which leads to waiting times for the loader while over-trucking reduces trucks utilisation which leads to trucks queue at the loader. The waiting times decreases the overall productivity of the haulage operations resulting into increased shovel-truck unit production cost, making the system more expensive. This research study was carried out in a limestone open pit mine at Mombasa Cement Limited (Vipingo plant) Kenya using queuing theory technique to study and optimize the haulage system. A multichannel queuing model (M1/M2/S/n: FCFS) was developed to capture the activities and predict the behaviour of the haulage system from loading at the shovel to dumping at the crusher and back at the loading points. The trucks inter-arrival time (min), service time (min), number of loaders, the truck capacities, and the current number of trucks in the system were recorded. This data was analysed based on the assumptions of a multichannel queuing approach with negative exponential inter-arrival time and negative exponential service time. The model was developed in Mat-lab software and used to calculate the inter-arrival rate and service rate for the different number of trucks subjected to the same queuing system. The current system gave an inter-arrival rate of 12 trucks/hour and service rate of 10 trucks/hour with 16 trucks and 2 servers in the system. Upon subjecting the data to the optimisation model, the results showed that when the number of trucks increased, the productivity of the shovel increased up to an optimal point after which, a iv further increase in the number of trucks reduced the truck productivity hence increasing cost per tonne hauled. The result indicated that the optimal fleet size was 12 trucks with 2 servers in operation and thus the 4 trucks could be sold or parked only to be used upon breakdown.