Given the dynamic nature of Urban Freight Transport (UFT) processes, the involved transport and logistics operators face with internal and external issues that should tackle to improve last-mile levels of service and decrease total costs while performing delivery operations. Customers (i.e., freight receivers) perceive the level of service through the acceptance of their requests, while total operational costs are mainly determined by the total travel costs (i.e., distance and/or time) required to accomplish the customers' request. In addition, the vehicle-kilometres travelled are related to the externalities produced. Given that the actors involved in the process operate in a stochastic environment (with changes that can occur both in terms of demand – receivers' requests, and in supply – travel times), collaboration and coordination among the operators could play a key role in meeting the customers' requests as well as in reducing both internal and external delivery costs. Therefore, the paper proposes an UFT modelling framework that integrates collaboration and coordination processes among the different involved actors, and allows the benefits to be assessed. The model has a multi-agent architecture based on microsimulation. In particular, the multi-agent architecture allows us to point out the different actors' responses to various internal (e.g., delivery requests) and external (e.g., delivery times) changes occurring in the daily delivery operations. It consists of three layers. The first one simulates the interactions among actors operating collaboratively. The second layer microsimulates the collaborative processes of information management. Finally, a third layer integrates the two previous layers, facilitating a decision-making process in such a dynamic context. The whole modelling framework is tested in a real case study in which it is possible to validate pros and cons of working in a collaborative and coordinative environment. The results show significant benefits from actors/operators involved in the process and subsequently can address the policy/measure implementation towards a more sustainable and liveable city. [ABSTRACT FROM AUTHOR]