We followed a two-step process:
- Impact analysis: The Tredence team put studied the inherent nature of sub-optimal ﬂeet transfers and understand its drivers. We were able to infer that 40% of the transfers occurred despite inventory availability, which was driven by anticipation of demand. However, about 35% of the transfers did not result in any rental in the short term, indicating inefﬁcient demand analysis.
- Solution development: We put together a demand forecast solution – a hybrid time series model that helped forecast short-term demand.
- The forecast model was scalable to predict demand at different levels, reducing suboptimal ﬂeet transfers
- It enabled optimized inventory planning for district manager, leading to better ﬂeet utilization