Traffic congestion and queues made by cars, are now part of our everyday life. Congestion levels have increased significantly in recent years, and this trend continues. Making correct decisions in congestion management is a difficult thing, because the decision factors need to analyze and absorb a large amount of information, which often are not relevant. In this paper, we propose a regression model to determine the quality factor used in road traffic. The model is defined starting from an experimental study aimed at analyzing the traffic quality in an urban network. This study represents a continuation of previous work [3], in which the authors proposed a new analytical model to determine the quality factor used in transport network (QF). The quality factor is measure used to characterize the global traffic in the intersection, but also to optimize traffic in a structure of interconnected intersections. This concept QF allows the use of the classical models of queuing theory to represent the road traffic. The experimental study was initiated in an area of urban transport network. Using the characteristics of the network topology we determined the value of quality factor for each connection represented in the studied network. With this study, we obtained information about the level of traffic congestion (utilization level), we could estimate the waiting time (in a queue) and response time (the elapsed time between the moment a traffic participant initiates an action and the time when he/she is allowed to execute the action).