System and process for job scheduling using limited discrepancy search

• US 6,070,144 A
• Filed: 08/27/1998
• Issued: 05/30/2000
• Est. Priority Date: 01/09/1996
• Status: Expired due to Term

• Current Assignee: University of Oregon
• Original Assignee: State of Oregon
• Inventors: Ginsberg, Matthew L., Crawford, James M., Harvey, William D., Jonsson, Ari K., Pemberton, Joseph C.
• Primary Examiner(s): Stamber, Eric W.
• Assistant Examiner(s): Patel, Jagdish
• Application Number:
  US09/141,249
• Time in Patent Office: 642 Days
• Field of Search: 705/7, 705/8, 705/9, 395/672, 395/674
• US Class Current: 705/7.17
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/063118 Staff planning in a project environment
• CPC Class Codes: G06Q 10/06312 Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
• CPC Class Codes: G06Q 10/06313 Resource planning in a project environment

Abstract

Assignment of attributes to elements subject to constraints is achieved using a system that has a systematic engine and a nonsystematic engine. The systematic engine includes a schedule developer for producing partial proposed assignments, a pruning processor for determining violations of discrepancy limits by a partial proposed assignment, and a bound selector for relaxing discrepancy limits as needed. The non-systematic engine includes a schedule packer for modifying assignments proposed by the systematic engine and an evaluator for comparing the modified assignments with the constraints.

Claims

1. A computer-implemented system for producing a schedule for a project by assigning attributes to elements in accordance with given constraints, the system comprising:

a systematic engine including;

a schedule developer adapted to generate an intermediate assignment of a subset of the attributes to a subset of the elements;

a pruning processor operatively coupled to the schedule developer and adapted to generate an error signal responsive to violation of a subset of predetermined discrepancy limits by the intermediate assignment, the error signal being applied to the schedule developer and causing the schedule developer to modify the intermediate assignment; and

a bound selector operatively coupled to the schedule developer and adapted to store and modify the discrepancy limits in response to operation of the schedule developer and the pruning processor; and

a nonsystematic engine, including;

a schedule packer operatively coupled to the systematic engine, adapted to accept as input the intermediate assignment and to produce therefrom an improvement thereof by a right-justify operation and a left-justify operation; and

an evaluator, operatively coupled to the schedule packer, adapted to estimate an overall resource utilization of the improvement, to generate as output a violation signal in response to the improvement violating a subset of the constraints, the violation signal being applied as an input to the systematic engine and causing the schedule developer to generate a new intermediate assignment, and to generate as output the improvement as the schedule for the project responsive to the improvement not violating the constraints.

2. A computer-implemented system as in claim 1, wherein the intermediate assignment includes an intermediate assignment start and an intermediate assignment end, and wherein:

(i) the left-justify operation includes choosing a left-justification start preceding the intermediate assignment start sufficiently to avoid overlap of assignments between a left-justification assignment and the intermediate assignment; and

(ii) the right-justify operation includes choosing a right-justification end succeeding the intermediate assignment end sufficiently to avoid overlap of assignments between a right-justification assignment and the intermediate assignment.

3. A computer-implemented system as in claim 1, wherein the right-justify operation and the left-justify operation are iterated.

4. A computer-implemented system as in claim 1, wherein the intermediate assignment includes an intermediate assignment end having a particular relationship to a cycle of shifts, the right-justify operation including selection of a right-justify end also having the particular relationship to the cycle of shifts.

5. A computer-implemented system as in claim 1, wherein the intermediate assignment includes an intermediate assignment end having a particular relationship to a cycle of shifts, the right-justify operation including selection of a right-justify end having a random relationship to the cycle of shifts.

6. A computer-implemented process for producing a schedule for a project by assigning attributes to elements, subject to a set of constraints, the process comprising:

a) generating a partial proposed assignment;

b) determining whether the partial proposed assignment violates a predetermined discrepancy limit;

c) modifying the partial proposed assignment responsive to violation of the predetermined discrepancy limit;

d) iterating b) and c) responsive to continued violation of the predetermined discrepancy limit and responsive to existence of additional possible proposed assignments for which violation of the predetermined discrepancy limit is not established;

e) altering the predetermined discrepancy limit and iterating a)-d) responsive to the non-existence of additional possible proposed assignments for which violation of the predetermined discrepancy limit is not established;

f) forming an intermediate assignment responsive to the partial proposed assignment;

g) modifying the intermediate assignment to produce an improvement of the intermediate assignment, using a right-justify operation and a left-justify operation;

h) evaluating the improvement with respect to the constraints;

i) iterating a)-h) responsive to the improvement violating a subset of the constraints; and

j) producing the schedule for the project from the improvement responsive to the improvement not violating the constraints.

7. A computer-implemented process as in claim 6, wherein the intermediate assignment includes an intermediate assignment start and an intermediate assignment end, and wherein:

(i) the left-justify operation includes choosing a left-justification start preceding the intermediate assignment start sufficiently to avoid overlap between a left-justification assignment and the intermediate assignment; and

(ii) the right-justify operation includes choosing a right-justification end succeeding the intermediate assignment end sufficiently to avoid overlap between a right-justification assignment and the intermediate assignment.

8. A computer-implemented process as in claim 6, wherein the right-justify operation and the left-justify operation are iterated.

9. A computer-implemented process as in claim 6, wherein the intermediate assignment includes an intermediate assignment end having a particular relationship to a cycle of shifts, the right-justify operation including selection of a right-justify end also having the particular relationship to the cycle of shifts.

10. A computer-implemented process as in claim 6, wherein the intermediate assignment includes an intermediate assignment end having a particular relationship to a cycle of shifts, the right-justify operation including selection of a right-justify end having a random relationship to the cycle of shifts.

11. A computer-implemented system for producing a schedule for a project by assigning attributes to elements in accordance with given constraints, the system comprising:

a systematic engine including;

a schedule developer adapted to generate an intermediate assignment of a subset of the attributes to a subset of the elements;

a pruning processor operatively coupled to the schedule developer and adapted to generate an error signal responsive to violation of a subset of predetermined discrepancy limits by the intermediate assignment, the error signal being applied to the schedule developer and causing the schedule developer to modify the intermediate assignment; and

a bound selector operatively coupled to the schedule developer and adapted to store and modify the discrepancy limits in response to operation of the schedule developer and the pruning processor; and

a nonsystematic engine, including;

a schedule packer operatively coupled to the systematic engine, adapted to accept as input the intermediate assignment and to produce therefrom an improvement thereof; and

an evaluator, operatively coupled to the schedule packer, adapted to estimate an overall resource utilization of the improvement, to generate as output a violation signal in response to the improvement violating a subset of the constraints, the violation signal being applied as an input to the systematic engine and causing the schedule developer to generate a new intermediate assignment, and to generate as output the improvement as the schedule for the project.

12. A computer-implemented system as in claim 11, wherein the bound selector modifies the discrepancy limits in response to detected inability of the systematic engine to generate new intermediate assignments without violating the discrepancy limits.

13. A computer-implemented system as in claim 11, wherein the schedule developer and the pruning processor iteratively operate together to generate partial assignments, and wherein each iteration of intermediate assignment generation results in addition of a new assignment of one of the attributes of one of the elements, in response to a previous iteration not resulting in generation of the error signal.

14. A computer-implemented system as in claim 11, wherein the constraints include a required deadline for completion of the project.

15. A computer-implemented system as in claim 11, wherein the discrepancy limits include a threshold number of violations of a predetermined heuristic.

16. A computer-implemented process for producing a schedule for a project by assigning attributes to elements, subject to a set of constraints, the process comprising:

a) generating a partial proposed assignment;

b) determining whether the partial proposed assignment violates a predetermined discrepancy limit;

c) modifying the partial proposed assignment responsive to violation of the predetermined discrepancy limit;

d) iterating b) and c) responsive to continued violation of the predetermined discrepancy limit and responsive to existence of additional possible proposed assignments for which violation of the predetermined discrepancy limit is not established;

e) altering the predetermined discrepancy limit and iterating a)-d) responsive to the non-existence of additional possible proposed assignments for which violation of the predetermined discrepancy limit is not established;

f) forming an intermediate assignment responsive to the partial proposed assignment;

g) modifying the intermediate assignment to produce an improvement of the intermediate assignment;

h) evaluating the improvement with respect to the constraints;

i) iterating a)-h) responsive to the improvement violating a subset of the constraints; and

j) producing the schedule for the project from the improvement responsive to the improvement not violating the constraints.

17. A computer-implemented process as in claim 16, wherein the generating a partial proposed assignment includes following a heuristic.

18. A computer-implemented process as in claim 16, wherein the discrepancy limits include a threshold number of violations of a predetermined heuristic.

19. A computer-implemented process as in claim 16, wherein the constraints include a required deadline for completion of the project.