The problem considered in this paper is to construct several routes and schedules of fleets that minimize the emission of a network consists of customers as nodes and their connection one to another by arcs, while varying the types of fleets and tightening the time windows. In the constructed models, the route of each fleet will be produced and treated as decision variable. The routes themselves depend on the capacity of each vehicle type and time windows constraints on each route. The objective of this paper is to construct routes of heterogeneous fleet problem with time windows with minimum emission, CO2eq in this context. A hybrid algorithm is used to construct the route. This algorithm is then validated using real demand historical data of a market in London. The result from model then used to analyze the differences which created by the algorithm and translated into a group of routes to be used.