||Traffic congestion in central business districts (CBD) has long been a headache for many metropolises worldwide. Hong Kong is no exception. In Hong Kong, the congestion is partly caused by the high number of bulky buses that enter the CBD, due to the adoption of the radial bus route networks. One of the possible solutions to this problem is merging some radial routes to form cross-town routes. This study aims to model the decision of merging radial routes and determine the frequencies of the merged routes so that a successful merging of routes will result, in terms of saving in the number of bus trips entering the CBD, the total number of buses used for serving the resultant bus route network, the difference between the two radial routes that are to be merged, and the degree of overlap of the two merged routes. Meanwhile, the merger is subjected to the constraints that the service will not be degraded in terms of frequency and the ease with which passengers can board the buses. A multi-objective mixed-integer non-linear program is formulated to model the problem and a decomposition heuristic is developed to solve the mathematical model. The model is tested by a set of data on the bus routes terminating in Admiralty, Central and Sheung Wan - the CBD of Hong Kong. From the results given by this set of data, it was found that the number of bus trips could be saved by about 6%. This 6% saving induces a 19% saving in bus stopping activity. These savings help reduce the traffic congestion in these areas. This study also analyzes the effects on merger results of different settings of parameters in the model. It is concluded that the greater the overlap between two radial bus routes, the more beneficial the merger will be.