Print Revenue Risk Assessment Using Dynamic Model Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managed print services face challenges in accurately determining print revenue behavior and associated risk levels, leading to uncertainties in cost prediction and customer dissatisfaction due to fluctuating print volumes and prices.
Innovation Solution
A system and method utilizing a computing device to determine print revenue for each time period, select a best-fit model from multiple models (linear, flat, first-order, and second-order autoregressive) based on historical data, and calculate a risk level by analyzing the difference between the model and actual revenue, enabling risk assessment and forecasting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple models are used to determine print revenue behavior, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system dynamically selects from multiple statistical models (linear, flat, first-order autoregressive, second-order autoregressive) based on which model best fits the actual print revenue data. This dynamic model selection allows the system to adapt to different revenue patterns while maintaining measurement precision without permanently increasing system complexity.
Solution Approach 2:
The system changes the parameter of model selection by evaluating different statistical models and choosing the one that best fits the observed print revenue behavior. This allows the system to maintain high measurement precision for risk level determination by selecting the most appropriate model for each specific account's revenue pattern.
2Reliability
If print revenue is tracked for multiple time periods, then reliability of risk assessment is improved, but loss of time increases
Solution Approach 1:
The system collects print revenue data across multiple time periods before performing risk assessment, ensuring that sufficient historical data is available to reliably determine revenue behavior patterns. This preliminary data collection phase establishes a solid foundation for accurate risk assessment while managing the time investment required.
Solution Approach 2:
The system uses historical print revenue data from multiple time periods as feedback to determine the account's revenue behavior and select the appropriate statistical model. This feedback mechanism improves reliability by base(ing) assessments on actual historical performance rather than assumptions.
3Productivity
If best fit model selection is implemented, then productivity of risk assessment is improved, but difficulty of detecting and measuring increases
Solution Approach 1:
The system segments the risk assessment process by dividing it into distinct steps: collecting print revenue data, selecting from predefined statistical models (linear, flat, first-order autoregressive, second-order autoregressive), fitting the selected model to the data, and determining risk level. This segmentation makes the complex model fitting process more manageable and efficient.
Solution Approach 2:
The system employs a universal approach by using multiple statistical models that can be applied to different types of print revenue patterns. This multi-functionality allows the same framework to handle various revenue behaviors (increasing, decreasing, stable, seasonal) without requiring completely different assessment methodologies.
Data Source
AI summary
Methods and systems for determining print revenue behavior for an account are disclosed. A computing device may determine a print revenue for an account for each of a plurality of time periods. The computing device may select a best fit model from a plurality of models based on the print revenue for the account for the plurality of time periods. The computing device may determine a risk level associated with the account based on a best fit model. The risk level may be based on a difference between the best fit model and the print revenue for the account for each of time period.