Scheduling a process according to time-varying input prices

• US 8,463,616 B2
• Filed: 10/15/2002
• Issued: 06/11/2013
• Est. Priority Date: 10/15/2002
• Status: Active Grant

• Current Assignee: Schneider Electric Systems USA, Inc. (Schneider Electric SE)
• Original Assignee: Invensys Systems Incorporated (Invensys PLC)
• Inventors: Fox, Brenton A., Strand, William C.
• Primary Examiner(s): Ruhl, Dennis
• Assistant Examiner(s): Fisher, Paul R
• Application Number:
  US10/271,365
• Publication Number:
  US 20040073438A1
• Time in Patent Office: 3,892 Days
• Field of Search: 705/1, 705/1.1, 705/80, 705300-348, 705901-912, 705/412
• US Class Current: 705/1.1
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 30/0283 Price estimation or determination
• CPC Class Codes: H02J 3/008 involving trading of energy or energy transmission rights
• CPC Class Codes: Y04S 10/50 Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
• CPC Class Codes: Y04S 50/10 Energy trading, including energy flowing from end-user application to grid
• CPC Class Codes: Y04S 50/14 Marketing, i.e. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards

Abstract

Disclosed are methods for scheduling when to run a process to take advantage of time-varying input prices by predicting the “rank” of the current actual input price among future input prices. The rank is that portion of a future time period during which the input price will be less than the current actual input price. An “operational cutoff” is set based on the portion of time during which a process should run in order to produce a target output amount. Periodically, the actual input price is determined, and its rank is predicted. The predicted rank is then compared with the operational cutoff. If the predicted rank is below the cutoff, then it makes economic sense to run the process. Otherwise, it would be cheaper not to run the process at present but to wait a while in expectation that the input price will drop.

Claims

1. A method carried out by a computer system including a physical computer-readable medium including computer-executable instructions for deciding whether to run a process to take advantage of time varying input commodity prices in order to reduce expenditures based on an actual price of an input commodity to the process while achieving an actual output amount at least as great as a target output amount, the actual output amount based, at least in part, on an amount of time that the process runs, the computer-executable instructions comprising:

determining a prediction time period for the process to run based on the target output amount;

determining a first portion of the prediction time period during which the input commodity is available for the process to run in order to achieve the target output amount;

determining a plurality of input commodity prices available for the process to run during the first portion of the prediction time period;

calculating a fractional rank for each of the input prices during the first portion of the prediction time period;

selecting an actual price of the input commodity based on a selected fractional rank if the process is run in a present time frame;

predicting a second portion of the prediction time period during which the price for the input commodity will be less than the selected actual price of the input commodity in the present time frame;

comparing the predicted second portion of the prediction time period with the determined first portion of the prediction time period; and

deciding to run the process in response to the predicted second portion of the prediction time period being less than the determined first portion of the prediction time period.

2. The method carried out by the computer system of claim 1 wherein predicting the second portion of the prediction time period comprises using a value selected from the group consisting of:

the selected actual input commodity price, actual input commodity price in a prior period, rank of the selected actual input commodity price in a set of actual input commodity prices over a prior period, rank of the selected actual input commodity price in a set of actual input commodity prices in a day of the week, change in the actual price of the commodity input over a prior period, and predicted demand of the input commodity.

3. The method carried out by the computer system of claim 1 wherein predicting the second portion of the prediction time period comprises:

predicting prices of the input commodity during the prediction time period;

dividing the prediction time period into sub-periods, the dividing consistent with the predicted input commodity price being substantially constant during each sub-period; and

assigning to the predicted second portion of the prediction time period during which the input commodity price will be less than the selected actual price of the input commodity having the selected fractional rank, a portion of the prediction time period represented by sub-periods with predicted input prices less than the selected actual price of the input commodity having the selected fractional rank.

4. The method carried out by the computer system of claim 3 wherein predicting prices of the input commodity comprises predicting absolute input commodity prices based on the fractional rank for each of the input commodity prices.

5. The method carried out by the computer system of claim 3 wherein predicting prices of the input commodity comprises predicting relative input commodity prices based on the fractional rank for each of the input commodity prices.

6. The method carried out by the computer system of claim 3 wherein dividing the prediction time period comprises dividing the prediction time period into equal-length sub-periods.

7. The method carried out by the computer system of claim 3 wherein dividing the prediction time period comprises dividing the prediction time period into contiguous sub-periods of maximal length consistent with the predicted input commodity price being substantially constant during each sub-period.

8. The method carried out by the computer system of claim 3 wherein dividing the prediction time period comprises assigning one predicted input commodity price to each sub-period.

9. The method carried out by the computer system of claim 3 wherein dividing the prediction time period comprises setting a price differential threshold and wherein the predicted input price is constant within a range defined by the price differential threshold for each sub-period.

10. The method carried out by the computer system of claim 3 wherein the input comprises a plurality of sub-inputs;

wherein predicting prices of the input commodity comprises predicting prices of a plurality of sub-inputs;

wherein dividing the prediction time period into sub-periods is consistent with a sum of the predicted sub-input prices being substantially constant during each sub-period and each price below the selected fractional rank; and

wherein assigning to the predicted second portion comprises assigning to the predicted second portion of the prediction time period during which the input commodity price will be less than the selected actual input commodity price, a portion of the prediction time period represented by sub-periods with sums of the predicted sub-input prices less than the selected actual input commodity price based on the selected fractional rank.

11. The method carried out by the computer system of claim 1 further comprising the computer-executed steps of:

determining a new first portion of a new prediction time period during which the process should run in order to achieve the target output amount;

determining a plurality of new input commodity prices available for the process to run during the new determined first portion of the new prediction time period;

calculating a new fractional rank for each of the new input commodity prices during the new determined first portion of the new prediction time period;

selecting a new actual price of the input commodity based on a selected new fractional rank if the process is run in a new present time frame;

predicting a new second portion of the new prediction time period during which the input commodity price will be less than the new actual price of the input commodity in the new present time frame;

comparing the new predicted second portion of the new prediction time period with the new determined first portion of the new prediction time period; and

if the new predicted second portion of the new prediction time period is less than the new determined first portion of the new prediction time period, then deciding to run the process in the present time frame.

12. The method carried out by the computer system of claim 11 further comprising the computer-executed steps of:

assigning to the target output amount a new value, the new value based, at least in part, on an amount of output already produced.

13. A non-transitory physical computer-readable medium containing instructions for performing a method for deciding whether to run a process in order to reduce expenditures on an input to the process while achieving an actual output amount at least as great as target output amount, a price of the input varying through time, the actual output amount based, at least in part, on an amount of time that the process runs, the method comprising:

determining a prediction time period for the process to run based on the target output amount to be produced;

determining a first portion of the prediction time period during which the process should run in order to achieve the target output amount;

calculating a fractional rank for each of the input prices during the first portion;

determining an actual price of the input based on a selected fractional rank if the process is run in a present time frame;

predicting a second portion of the prediction time period during which the input price will be less than the determined actual price of the input in the present time frame;

comparing the predicted second portion of the prediction time period is less than the determined first portion of the prediction time period, then deciding to run the process in the present time frame.

14. The physical computer-readable medium of claim 13 wherein predicting comprises using a value selected from the group consisting of:

the selected actual input price, actual input price in a prior period, rank of the selected actual input price in a set of actual input prices over a prior period, rank of the selected actual input price in a set of actual input prices in a day of the week, change in the actual price of the input over a prior period, and predicted demand of the input.

15. The physical computer-readable medium of claim 13 wherein predicting comprises:

predicting prices of the input during the prediction time period;

dividing the prediction time period into sub-periods, the dividing consistent with the predicted input price being substantially constant during each sub-period; and

assigning to the predicted second portion of the prediction time period during which the input price will be less than the determined actual price of the input based on the selected fractional rank, a portion of the prediction time period represented by sub-periods with predicted input prices less than the determined actual price of the input.

16. The physical computer-readable medium of claim 15 wherein predicting prices of the input comprises predicting absolute input prices based on the fractional rank.

17. The physical computer-readable medium of claim 15 wherein predicting prices of the input comprises predicting relative input prices based on the fractional rank.

18. The physical computer-readable medium of claim 15 wherein dividing the prediction time period comprises dividing the prediction time period into equal-length sub-periods.

19. The physical computer-readable medium of claim 15 wherein dividing the prediction time period comprises dividing the prediction time period into contiguous sub-periods of maximal length consistent with the predicted input price being substantially constant during each sub-period.

20. The physical computer-readable medium of claim 15 wherein dividing the prediction time period comprises assigning one predicted input price to each sub-period.

21. The physical computer-readable medium of claim 15 wherein dividing the prediction time period comprises setting a price differential threshold and wherein the predicted input price is constant within a range defined by the price differential threshold for each sub-period.

22. The physical computer-readable medium of claim 15 wherein the input comprises a plurality of sub-inputs;

wherein predicting prices of the input comprises predicting prices of a plurality of sub-inputs;

wherein dividing the prediction time period into sub-periods is consistent with a sum of the predicted sub-input having a price below the selected fractional rank and the predicted sub-input prices being substantially constant during each sub-period; and

wherein assigning to the predicted second portion comprises assigning to the predicted second portion of the prediction time period during which the input price will be less than the determined actual input price, a portion of the prediction time period represented by sub-periods with sums of the predicted sub-input prices less than the determined actual input price based on the selected fractional rank.

23. The physical computer-readable medium of claim 13 further comprising:

determining a new first portion of a new prediction time period during which the process should run in order to achieve the target output amount;

determining a plurality of new input prices available for the process to run during the new prediction time period;

calculating a new fractional rank for each of the new input prices during the new first portion;

selecting a new actual price of the input based on a selected new fractional rank if the process is run in a new present time frame;

predicting a new second portion of the new prediction time period during which the input price will be less than the new actual price of the input in the new present time frame;

comparing the new predicted second portion of the new prediction time period with the new determined first portion of the new prediction time period; and

if the new predicted second portion of the new prediction time period is less than the new determined first portion of the new prediction time period, then deciding to run the process in the present time frame.

24. The physical computer-readable medium of claim 23 further comprising:

assigning to the target output amount a new value, the new value based, at least in part, on an amount of output already produced.

25. A method carried out by a computer system including a physical computer-readable medium including computer-executable instructions for deciding to run a process in order to reduce expenditures on an input to the process while achieving an actual output amount at least as great as a target output amount, a price of the input varying through time, the actual output amount based, at least in part, on an amount of time that the process runs, the method comprising the computer-executed steps of:

determining a first amount of time within a prediction period during which the process should run in order to achieve the target output amount;

determining a plurality of input commodity prices available for the process to run during the first amount of time;

calculating a fractional rank for each of the input prices during the first amount of time;

selecting an actual price of the input based on a selected fractional rank if the process is run in a present time frame;

predicting a second amount of time of the prediction period during which the price of the input will be less than the selected actual price of the input in the present time frame;

comparing the predicted second amount of time within the prediction period with the determined first amount of time; and

if the predicted second amount of time within the prediction period is less than the determined first amount of time within the prediction period, then deciding to run the process in the present time frame.

26. A non-transitory physical computer-readable medium containing instructions for performing a method for deciding to run a process in order to reduce expenditures on an input to the process while achieving an actual output amount at least as great as a target output amount, a price of the input varying through time, the actual output amount based, at least in part, on an amount of time that the process runs, the method comprising:

determining a first amount of time within a prediction period during which the process should run in order to achieve the target output amount;

determining a plurality of input commodity prices available for the process to run during the first amount of time;

calculating a fractional rank for each of the input prices during the first amount of time;

selecting an actual price of the input based on a selected fractional rank if the process is run in the present time frame;

predicting a second amount of time of the prediction period during which the input price will be less than the selected actual price of the input in the present time frame;

comparing the predicted second amount of time within the prediction period with the determined first amount of time; and

if the predicted second amount of time within the prediction period is less than the determined first amount of time within the prediction period, then deciding to run the process in the present time frame.