Method and apparatus for utility pricing analysis
View Patent ↗In one implementation, a volatility of expected service usage is calculated along with an expected revenue. User input provides estimates of service usage such as an estimated high usage, an estimated low usage, and a swing range of percentiles for the high and low usage estimates, for each time period in a plurality of future time periods. A user also provides a schedule of prices that are based on usage rates. A model volatility is calculated based on the swing range of percentiles, the estimated high and low usage, and a cumulative density function, such as a cumulative lognormal density function, for each time period in a plurality of time periods. The expected usage and the schedule of prices, along with any discount plans, are also used to calculate an estimated revenue.
1. A method in a computer for providing a pricing algorithm for a device, the device being a hardware resource or a software application, comprising:
receiving by the computer estimates of service usage for the device including a history of past usage of the device;
receiving by the computer a usage-based schedule of prices for the device; and
in response to the receiving the estimates of service usage and the receiving the schedule of prices:
calculating by the computer a model volatility of service usage relating to statistical variation in expected service usage of the device, the calculating of the model volatility including
averaging logarithms of past usage values as indicated by the history of past usage of the device, and
calculating a variance of the logarithms of the past usage values, and
calculating by the computer a revenue estimate for the usage-based schedule of prices that factors in discounts by looping over periods of usage and performing:
setting a price for the period to a base price;
determining average usage for the period; and
when the average usage for the period exceeds a discount usage, applying a discount to reset the price.
2. The method of claim 1 , wherein:
the estimates of service usage comprise an estimated high usage, an estimated low usage, and a swing range of percentiles; and
the calculating the model volatility comprises:
calculating a first volatility based on the swing range of percentiles, the estimated high usage, and a cumulative density function,
calculating a second volatility based on the swing range of percentiles, the estimated low usage, and a cumulative density function, and
calculating the model volatility based on the first and second volatilities.
3. The method of claim 2 , wherein the cumulative density function is a cumulative lognormal density function.
4. The method of claim 1 , further comprising:
receiving a set of discount rules; and
the calculating the revenue estimate for the usage-based schedule of prices is further performed in response to the set of discount rules.
5. The method of claim 4 , wherein the discount rules comprise rules based on one or more of: usage history or usage thresholds.
6. The method of claim 1 , further comprising:
receiving a set of anchor values representing usage-independent revenue rates; wherein
the calculating the revenue estimate for the usage-based schedule of prices is based on the anchor values.
7. The method of claim 1 , wherein
the estimates of service usage comprise an estimated high usage, an estimated low usage, and a swing range of percentiles, for each time period in a plurality of time periods; and
the calculating the model volatility comprises:
calculating a first volatility based on the swing range of percentiles, the estimated high usage, and a cumulative density function, for each time period in a plurality of time periods,
calculating a second volatility based on the swing range of percentiles, the estimated low usage, and a cumulative density function, for each time period in a plurality of time periods, and
calculating the model volatility based on the first and second volatilities, for each time period in a plurality of time periods.
8. The method of claim 7 , wherein the cumulative density function is a cumulative lognormal density function.
9. A computer system for providing a pricing algorithm for a device, the device being a hardware resource or a software application, comprising:
a memory for storing computer-executable instructions that implement:
a component that receives estimates of service usage for the device including a history of past usage of the device;
a component that receives a usage-based schedule of prices for the device; and
a component that, in response to the receiving the estimates of service usage and the receiving the schedule of prices:
calculates a model volatility of service usage relating to statistical variation in expected service usage of the device, the calculating of the model volatility including
averaging logarithms of past usage values as indicated by the history of past usage of the device, and
calculating a variance of the logarithms of the past usage values, and
calculates a revenue estimate for the usage-based schedule of prices that factors in discounts by looping over periods of usage and performing:
setting a price for the period to a base price;
determining average usage for the period; and
when the average usage for the period exceeds a discount usage, applying a discount to reset the price; and
a processor for executing the computer-executable instructions stored in the memory.
10. The system of claim 9 , wherein:
the estimates of service usage comprise an estimated high usage, an estimated low usage, and a swing range of percentiles; and
the component that calculates the model volatility comprises:
a component that calculates a first volatility based on the swing range of percentiles, the estimated high usage, and a cumulative density function,
a component that calculates a second volatility based on the swing range of percentiles, the estimated low usage, and a cumulative density function, and
a component that calculates the model volatility based on the first and second volatilities.
11. The system of claim 10 , wherein the cumulative density function is a cumulative lognormal density function.
12. The system of claim 9 , further comprising:
a component that receives a set of discount rules; and
the component that calculates the revenue estimate for the usage-based schedule of prices is configured to operate in response to the set of discount rules.
13. The system of claim 12 , wherein the discount rules comprise rules based on one or more of: usage history or usage thresholds.
14. The system of claim 9 , further comprising:
a component that receives a set of anchor values representing usage-independent revenue rates; wherein
the component that calculates the revenue estimate for the usage-based schedule of prices is configured to operate in response to the anchor values.
15. The system of claim 9 , wherein:
the estimates of service usage comprise an estimated high usage, an estimated low usage, and a swing range of percentiles, for each time period in a plurality of time periods; and
the component for calculating the model volatility comprises:
a component for calculating a first volatility based on the swing range of percentiles, the estimated high usage, and a cumulative density function, for each time period in a plurality of time periods,
a component for calculating a second volatility based on the swing range of percentiles, the estimated low usage, and a cumulative density function, for each time period in a plurality of time periods, and
a component for calculating the model volatility based on the first and second volatilities, for each time period in a plurality of time periods.
16. A computer-readable storage medium storing computer-executable instructions for execution by one or more processors for providing a pricing algorithm for a device, the device being a hardware resource or a software application, by performing a method comprising acts of:
receiving estimates of service usage for the device including a history of past usage of the device;
receiving a usage-based schedule of prices for the device; and
in response to the receiving the estimates of service usage and the receiving the schedule of prices:
calculating by a processor a model volatility of service usage relating to statistical variation in expected service usage of the device, the calculating of the model volatility including
averaging logarithms of past usage values as indicated by the history of past usage of the device, and
calculating a variance of the logarithms of the past usage values, and
calculating by a processor a revenue estimate for the usage-based schedule of prices that factors in pricing discounts.
17. The computer-readable storage medium of claim 16 , wherein:
the estimates of service usage comprise an estimated high usage, an estimated low usage, and a swing range of percentiles, for each time period in a plurality of time periods; and
the calculating of the model volatility comprises:
calculating a first volatility based on the swing range of percentiles, the estimated high usage, and a cumulative density function, wherein the cumulative density function is a cumulative lognormal density function, for each time period in a plurality of time periods,
calculating a second volatility based on the swing range of percentiles, the estimated low usage, and a cumulative density function, for each time period in a plurality of time periods, and
calculating the model volatility based on the first and second volatilities, for each time period in a plurality of time periods.