|Abstract||The project was initiated to investigate techniques to improve the long-term viability of the Mossman sugar-producing area through better harvesting techniques. The overall aim was to dovetail with previous SRDC-funded research by Mossman Mill that had suggested that a cane supply much lower in extraneous matter could potentially eliminate the need for expansion of mill capacity, negating a capital expenditure of at least $9 million dollars. The major brief of this project was to help implement these techniques, whilst ensuring that they did not cause any additional sucrose loss in the field.Due to varying crop conditions throughout the lifecycle of the project, it was impossible to gauge whether there had been any net decrease in the level of extraneous matter in the cane supply. Nonetheless, the project did provide some significant quantifiable benefits to the Mossman sugar-producing region, not the least of which was the further development and widespread adoption of a process called ?Feed-Train Optimisation? which has been shown to reduce sucrose loss and improve billet quality during mechanical harvesting.This project also undertook some novel research into the linkages between billet length and productivity, which indicated a previously unknown linkage between billet length and declining CCS in far north Queensland. The results suggest that a decrease in billet length of just 20 mm can decrease the CCS of green cane by approximately half a unit, and increase extractor losses by 5-10%. These results are considered to be extremely relevant to the current trend of CCS decline in far north Queensland.Trials investigating the effects of billet length upon productivity highlighted another previously unrecorded phenomenon, that average billet length changes with fan speed, and that there is an interaction between this change and the nominal billet length setting of the harvester. This may have implications upon the interpretation of previous research and the future design of machinery.Technological outputs of this research included a computer model of the feed-train and chopper system of modern mechanical harvesters, and the development of a computer system designed to give harvester operators a more intuitive guide to their performance relative to their peers. These are being used commercially.