Method and apparatus for queue-based automated staff scheduling

• US 7,246,076 B2
• Filed: 12/09/2004
• Issued: 07/17/2007
• Est. Priority Date: 12/09/2003
• Status: Expired due to Fees

• Current Assignee: Lightning Bolt Solutions, Inc. (PerfectServe, Inc.)
• Original Assignee: Lightning Bolt Solutions, Inc. (PerfectServe, Inc.)
• Inventors: Vajracharya, Suvas, Setiawan, David, Tjahjono, Robby
• Primary Examiner(s): Jeanty; Romain
• Application Number:
  US11/010,045
• Publication Number:
  US 20050125278A1
• Time in Patent Office: 950 Days
• Field of Search: 705/8, 705/9
• US Class Current: 705/7.21
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/063112 Skill-based matching of a person or a group to a task
• CPC Class Codes: G06Q 10/063116 Schedule adjustment for a person or group
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities
• CPC Class Codes: G06Q 10/109 Time management, e.g. calendars, reminders, meetings or time accounting
• CPC Class Codes: G06Q 10/1097 Task assignment

Abstract

A queue-based scheduling system, which comprises an automated staff scheduling computer program that is highly flexible in enforcing scheduling rules. This flexibility comes from the ability to: (1) Define conditional and unconditional rules; (2) Rank the rules/requests in varying priority as represented by a numeric value assigned to each rule/request; and (3) Specify rules both per individual and per group. These three abilities synergistically produce an automatic scheduling system that can enforce a wide variety of scheduling rules and requirements seen in actual staff scheduling situations. Furthermore, all these abilities rely on using a queue per scheduled assignment to hold requests and rules.

Claims

1. A queue-based scheduling apparatus for automated staff scheduling and for enforcing scheduling rules and requirements, comprising:

means for defining conditional and unconditional rules;

means for ranking rules and requests in varying priority as represented by a numeric value assigned to each rule and request; and

means for specifying rules both per individual and per group; and

means for generating schedules in multiple phases;

wherein during each phase, an automatic scheduler entirely schedules every assignment on every day in a scheduling period;

wherein on second and later phases, said auto scheduler learns from any hazards detected in earlier phases to generate a better overall schedule;

wherein said hazards are detected whenever said automatic scheduler cannot find an individual that would not violate some rule or request;

said automatic scheduler determining if any conditional rule has caused detection of hazard, and if said conditional rule is satisfied, finding a source or a condition of the rule that generated a violating request;

said automatic scheduler generating and submitting an internal request; and

in subsequent phases, said automatic scheduler avoiding said hazard by using said submitted request from previous phases.

2. The apparatus of claim 1, further comprising:

at least one queue per each scheduled assignment for holding requests and rules.

3. A scheduling process, comprising the steps of:

collecting requests and rules using a graphical user-interface;

determining a scheduling period that has a user-defined start date and stop date;

creating a request queue for each day of a period;

for each request or rule, determining which days said request or rule is applicable;

for each applicable day, inserting a request into its queue;

sorting said days from most constrained to least constrained;

beginning scheduling from a most constrained day;

for each assignment in a day, creating a list of candidates for that assignment for that day;

for each candidate, associating a highest priority request in said request queue that applies to said candidate, wherein said highest priority value for a candidate is a candidate bid for a chosen assignment;

sorting said candidates according to a bid calculated in the immediately preceding step;

scheduling a candidate having a highest number;

determining a schedule for any remaining days; and

means for generating schedules in multiple phases;

wherein during each phase, an automatic scheduler entirely schedules every assignment on every day in a scheduling period;

wherein on second and later phases, said automatic scheduler learns from any hazards detected in earlier phases to generate a better overall schedule;

wherein said hazards are detected whenever said automatic scheduler cannot find an individual that would not violate some rule or request;

said automatic scheduler determining if any conditional rule has caused detection of hazard, and said conditional rule is satisfied, finding a source or a condition of the rule that generated a violating request;

said automatic scheduler generating and submitting an internal request; and

in

subsequent phases, said automatic scheduler avoiding said hazard by using said submitted request from previous phases.

4. The method of claim 3, wherein said constraints on scheduling choices comprise either of conditional and unconditional constraints, wherein conditional constraints are only in effect if a certain scheduling condition applies.

5. The method of claim 3, wherein a rule applies to either of a specific individual or a group of individuals.

6. The method of claim 3, further comprising the step of:

associating rules with a number which describes a rank or a priority of that rule relative to other rules;

wherein a large positive number indicates that any individual under that rule is highly favored to be scheduled for the given shift on a given day;

wherein a large negative number indicates that any individual under that rule is not favored to be scheduled for the given shift on a given day; and

wherein an individual who is not subject to any rules has a priority point zero by default, indicating neutrality.

7. The method of claim 3, further comprising the step of:

choosing a rule having a highest absolute value of priority points for an individual to compete with other individuals when more than one rule is to be applied.

8. The method of claim 3, further comprising the steps of:

entering static or unconditional rules in the request queue prior to running the automated scheduler; and

entering conditional rules and requests during execution of the steps of said method, depending on the scheduling conditions;

wherein a highly adaptive scheduler is provided that changes its scheduling behavior during execution of the method, based on scheduling decisions made during execution of the method.

9. The method of claim 3, further comprising the step of:

providing conditional rules, wherein each conditional rule has a set of conditions and a set of actions, wherein when all conditions are met, said set of actions is converted into a request.

10. The method of claim 3, further comprising the step of:

providing an alternates list that exposes information in a form a user can use to override the choice that said automatic scheduler has made;

wherein said alternates list displays all constraints and requests that are currently imposed on alternate individuals.

11. The method of claim 3, further comprising the steps of:

said auto scheduler randomly choosing different assignments on different days, rescheduling individuals until a global schedule improves or converges to an acceptable one; and

said auto scheduler summarizing a number of requests or rules in each queue that are in violation across all assignments in an entire schedule to determine whether a global schedule has improved after a scheduling adjustment;

wherein if a total number of violations decreases, said schedule has been improved and therefore is chosen over prior schedules.