Outlier Detection package for service parts usage analysis
This is the package used for detect abnormal usage data of service parts in supply chain and logistics. The abnormaly is justified by comparing the install base data with the usage data when the trends of these two are similar. when dissimilar exists between these two, the trend of usage itself is used.
User recommendations
- The input data of Install base and Usage should have two columns: the first column should be the date format and the second should be numerical.
- The starting date of Install base should be earlier than the one of Usage and the ending date of Install base should be later than the one of Usage.
- The output of the package is a) plot (res$plot) and b) outliers (res$anoms)