This thesis proposes a traffic signal optimization method for balancing the interests of cyclists and motorized traffic. Traditionally, in bicycle traffic has not been given the same priority as motorized traffic in the signalization system planning. As a consequence, cyclists have to stop many times during their journey, which leads to a waste of energy and causes inconvenience, and at the same time experience long waiting times. In other words, stopping for a red light is a primary discouragement to cycling, since it costs a great deal of extra time and energy. This research relies on this ascertainment and focuses on designing a system which takes into account both types of road users, cyclists and motorized vehicles. Specifically, the proposed system receives the required data from the network, speeds of motorized vehicles and position of both road users, and based on a multi-objective optimization changes the green times of the traffic lights and gives comfortable speed advice to cyclists in order to cross the intersection without having to stop. For the aforementioned optimization the various objectives, namely delay of motorized vehicles and delay, number of stops and deviation between the desired and the advised speed of cyclists, are formulated and are combined as a weighted sum, as a function of green times and individual speed advice. The delay of both road users is formulated as the sum of the delay because of the traffic signal, the delay because of the queues formed at the intersection and the overflow delay, i.e. the additional delay caused when the arrival rate is greater than the service rate at the traffic signal. In regard to the number of stops, each cyclist receives a cost, every time that he has to stop. Last but not least, each cyclist has a desired speed and deviations of it come at a cost. This cost is higher for speeds higher than the desired speed than for speeds lower than the desired speed. The optimization is performed individually in each intersection when cyclists gets green. The time horizon length is set to 10 cycles and the first cycle of the resulting plan is implemented, before the computation is repeated. The cyclists and the motorized vehicles who arrive beyond horizon, receive a very high cost. Different scenarios are run in order to test the proposed system. In all scenarios, the average delay and the average number of stops of cyclists decrease, while the average delay and the average number of stops of motorized vehicles increase. The first two scenarios are used for the controller’s tuning and offer the first insight of the system. In the following two scenarios, the bicycle green wave is investigated. In the last two scenarios, two different components of the system, the number of stops of cyclists and the comfort of the speed advice are examined. Specifically, in the fifth scenario, the coefficient defining the relative importance of cyclist’s number of stops is investigated. Regardless of the value of the coefficient, the reduction of the average number of stops of cyclists is almost the same, with the highest value 69%. In the sixth scenario, the coefficient defining the relative importance of the deviation between the desired and the advised speed is examined. The higher the value of the coefficient, the smaller the deviation between the desired and the advised speed. Although, traditionally not much attention is paid to cyclists in the initial adjustment of the traffic light, the proposed traffic signal optimization method takes into account both types of road users, motorized traffic and cyclists. Reduction in the average delay and the average number of stops of cyclists can be achieved, however has as a consequence the increase of the average delay and the average number of stops of motorized traffic. The same is observed about the comfort of the speed advice. In conclusion, it is possible to improve the situation for cyclists, regarding their delay, number of stops and the comfort of their advised speed, however this improvement comes at a cost for motorized traffic.