Urban population and expansion continue to rise, as have emergencies, and emergency response times are becoming longer in the city. Response times are critical in urban emergency management. Traditional navigation systems do not consider the preferential road usage rights and requirements crucial for emergency teams such as ambulances, fire departments, and police. In traditional navigation systems, features such as safety lanes and road width are not incorporated as routing criteria. Past emergency navigation studies do not contain many variables and this causes delays in emergency response processes. As a result, applications that process real-time traffic information and provide guidance according to different vehicle types are needed for emergency management.. This study aims to develop an application capable of routing for different vehicles according to criteria such as working hours, holidays, weather conditions, and seasonal effects. The application is developed with a novel approach in this context using machine learning methods and produces alternate routes with GPS navigation based on the real-time traffic situation. Moreover, the software also provides routing solutions for trucks, freight distribution, and automobiles and recommends unique routes for emergency vehicles based on factors including safety lanes, road width, corner turning angles, and right of way. The expected outcomes upon implementing this application include easy and fast access to the emergency locations, reduced response times, and improvement in urban emergency management services.