Categorizing Freeway Flow Conditions by Using Clustering Methods

Three pattern recognition methods were applied to classify freeway traffic flow conditions on the basis of flow characteristics. The methods are K-means, fuzzy C-means, and CLARA (clustering large applications), which fall into the category of unsupervised learning and require the least amount of knowledge about the data set. The classification results from the three clustering methods were compared with the Highway Capacity Manual (HCM) level-of-service criteria. Through this process, the best clustering method consistent with the HCM classification was identified. Clustering methods were then used to further categorize oversaturated flow conditions to supplement the HCM classification. The clustering results supported the HCM's density-based level-of-service criterion for uncongested flow. In addition, the methods provide a means of reasonably categorizing oversaturated flow conditions, which the HCM is currently unable to do.