| 摘要 |
<p>PROBLEM TO BE SOLVED: To obtain information regarding a maximum value including no unexpected variation by finding the product of the maximum value of a similarity- dimensional variation pattern of traffic intensity in a past cycle time and the similarity coefficient of the latest traffic intensity to the similarity-dimensional variation pattern from traffic intensity by cycle times at specific equal intervals and time sections obtained by dividing the cycle time by (n). SOLUTION: An observed traffic data base part 2 gathers various traffic intensity of a network 1 by days and times by cycle times having specific continuous equal intervals, i.e., days wherein cycle variation is considered. A similarity-dimensional variation pattern generation part 3 finds the pattern of traffic intensity within arbitrary past days, a time series data conversion part 4 finds the similarity coefficient of the traffic intensity of the days to the pattern, and a maximum extraction part 5 finds the maximum value of the pattern. A predicted traffic output part outputs the product of the similarity coefficient and maximum value as a maximum traffic predicted value.</p> |
