An Enterprise Revenue Management and Pricing System (ERMPS) is a long-term and sustainable company commitment. It increases the sophistication and personalization of Revenue Management, pricing practices, implements state of the art optimization and forecasting algorithms, and provides seamless deployment of inventory and pricing controls.
If your company has decided to build or update an Enterprise Revenue Management and Pricing System (ERMPS), defining sources of uncertainty and their treatment is a key design consideration when identifying the optimal balance between system stability and solution flexibility.
Typical sources of uncertainty that need to be considered when determining the solution for an Enterprise Revenue Management Pricing System are:
- Volatile data conditions arise when expected data inputs are either missing, insufficient or bad. Uncertainty treatment for volatile data conditions should focus on building preventive alerts that identify missing or insufficient data inputs and executing sanity checks that verify data integrity.
- Forecast error is an inherent source of uncertainty and represents the extent to which the model prediction (signal) is “similar” to the actual realization of demand and cannot be minimized beyond natural variation of demand (noise). Uncertainty treatment for forecast error should focus on improving the forecast signal-to-noise ratio by using pick-best forecasting methods, blending on-the-books, and incorporating effects of inventory controls and pricing.
- Model misspecifications are due to unwarranted model assumptions, not considering edge cases, and not incorporating business rules. Uncertainty treatment for unwarranted model assumptions is addressed by constructing prototypes that collect business user feedback at each stage of the Enterprise Revenue Management and Pricing System design process. Edge cases should be addressed by system wide defaulting strategies and parameters. Finally, to the extent that model infeasibilities or local optima are controlled, business rules should be incorporated into the model formulation.
- Environmental disruptions are drastic changes in your company’s supply and pricing strategies and/or your industry landscape. While uncertainty treatment for environmental disruptions is a larger topic worth further consideration, minimizing uncertainty associated with environmental disruptions should focus on incorporating adaptive gains as part of the forecast model, using flexible segmentation, periodic re-calibration of market response models, and building what-if capabilities.
While treatment of uncertainty conditions should emphasize system stability over solution flexibility, the optimal configuration for treatment of uncertainty is company specific and should be tailored according to the business requirements.
If your company is considering an Enterprise Revenue Management and Pricing System ask questions about: volatile data conditions, forecast error, model misspecifications and environmental disruptions and how they are dealt with, quantifying the robustness of the proposed solution to uncertainty.