Based on optimization setup technology, an online adaptive calculation method for improving accuracy of strip coiling temperature control on the run-out table (ROT) has been developed and implemented in hot strip mill (HSM). Multi-objective control strategies, which include coiling temperature, middle target temperature and appropriate cooling rates have been finalised. Cooling strategies, elements tracking, and dynamic correction are employed in the control system. In addition, the model optimization and soft-measure method are also introduced in the study. Rolling tests with various grades of steel covering a wide range of thickness show that the developed model can improve the accuracy of coiling temperature control to obtain an uniform mechanical properties. Good correlation has been found between the predicted temperatures and the actual coiling ones.