Assess the traffic situation will be very good for finding the fastest route or avoid the traffic jam. In order to do so, we need to know the traffic situation on all the roads. Speed of the flow of vehicles on the road can be measured by number of vehicles go by a check point within a unit of time. This measurement shows the status of that road, a low value points out a jam and a high value let us know the road is clear. But in the case there are no private lanes for every kind of vehicles and vehicles are moving straggling, counting vehicles is difficult. Another method is monitoring the flows though camera traffic and make evaluation of the status. This also requires more staff and the evaluation is not continuously nor homogeneous. Nowadays, all buses are equipped with GPS real-time trackers which send their data of speed, location to the center control. In the case buses is moving together with other vehicles on a busy road, we assume that buses having an average speed that could represent the movements of other vehicles. Within a specific road, there’re many buses go by within a period of time, making a specific set of speed data. From these sets, we can estimate average speed of this road. We also calculate the reliability of that result with some statistical functions. Furthermore, a huge amount of data sending to the center control is difficult for data processing. The system need to be designed in parallelism architecture for performance improving. The data will be divided by a receiver before they are sent to their sub-system for the estimation. After all, data collected day by day could be used to verify the current day’s calculating. So that, an effective database design is the last but not least. Moreover, the data collected day by day will provide the behavior of the flow on a road. Which region usually have jam or when the roads will be more heavier can be known by review the estimate result in the past. So that, the GPS data from buses will be more useful if they are collected and processed together. This paper shows how to collect speed data from buses, make sets of data and from that estimate the status of roads. By comparison with other methods, we also verify the GPS data from buses is useful or not.