- DIGITAL SOURCING
The solution recommends alternate route options for a journey in case confirmed tickets are unavailable on direct trains. An algorithm based approach helps provide options to customers from under-utilized trains in the network and hence improves overall capacity utilization of the existing trains, while satisfying incremental demand. Factors such as weather patterns, expected train delays, etc. are factored in while recommending alternate routes with connecting trains. The user has the flexibility of specifying layover duration cutoffs, type of trains, etc.
The solution focuses on improving the seat quota allocation across stations based on observed demand patterns. This involves quota optimization keeping in mind changing demand patterns across routes, seasons, etc. The quota optimization involves redistribution of seat quota and also recommendations around adding new stations for quota allocation.
The solution uses a simulation approach to show the impact of varying price points on demand and thus net revenue realization. Scenarios can be built for a particular class on a particular train for a given journey date. The solution considers the present utilization and simulates the expected utilization based on the price and observed booking patterns till date.
The solution provides recommendations around adding new coaches in a particular train based on expected cumulative demand till the journey date. A statistical analysis was done on historical data. Based on the observed booking pattern of a train (booking pattern from 4 months before journey date till journey date), a growth curve was fitted to model the expected booking pattern. These booking curves will vary for different train types and different source and destination pairs. The booking curve may also vary by time (trains with journey dates during festive periods or holidays will have a different booking curve). Based on the observed booking till a time point and extrapolating the expected booking till the journey date, using the booking curves derived, the solution will recommend adding or dropping coaches from a certain route. Recommendations will factor in constraints such as minimum and maximum number of coaches in a train.
The solution projects weekly revenue for a future horizon for any train/type of train/origin-destination combinationsby using various forecasting methodologies such as regression based models, time series models, etc., factoring in demand peaks and troughs such as those caused by seasonality/festive occasions.