Collaborative scheduling

• US 10,796,284 B2
• Filed: 09/20/2016
• Issued: 10/06/2020
• Est. Priority Date: 09/20/2016
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Mirzazad Barijough, Sanam, Malur Srinivasan, Ramya
• Primary Examiner(s): Miller, Alan S
• Assistant Examiner(s): Wang, Jay-Ming
• Application Number:
  US15/271,032
• Publication Number:
  US 20180082259A1
• Time in Patent Office: 1,477 Days
• Field of Search: 705 711- 742
• CPC Class Codes: G06Q 10/063116 Schedule adjustment for a person or group
• CPC Class Codes: G06Q 10/1093 Calendar-based scheduling for persons or groups

Abstract

A method of collaborative scheduling includes receiving group inputs and receiving environmental data. The method includes calculating collaborative schedule information based on the environmental data, traits, and the group inputs that include feasible schedules and total schedule costs associated therewith that are based on an individual member performance cost, a flexible task violation cost, and a behavior cost. The method includes receiving a selection of a feasible schedule as a preferred schedule and determining a first trait and a second trait of each member based thereon. The first trait includes selfishness assumed when the feasible schedule selected places a higher total schedule cost on other members. The second trait includes selflessness assumed when the feasible schedule selected places a lower total schedule cost on the other members. The method includes generating a collaborative schedule as the feasible schedule selected by the members.

Claims

1. A method of collaborative scheduling, the method comprising:

setting, by a calendar server, an initial value for at least one of a selflessness trait or a selfishness trait of each member of a plurality of members who are included in a group to which a collaborative schedule pertains, the at least one of the selflessness trait or the selfishness trait related to behavior of a given member as indicated by preferences of the given member;

receiving, at the calendar server, group inputs that include individual member tasks, mutual member tasks, individual member constraints, and mutual member constraints;

receiving, at the calendar server, environmental data collected by a data collection engine;

calculating, by the calendar server, first collaborative schedule information based on the environmental data, the initial value for the at least one of the selflessness trait and the selfishness trait, and the group inputs, the first collaborative schedule information including a number of first feasible schedules for a first task, event, or responsibility and total schedule costs that are associated with the first feasible schedules for each member, the total schedule costs for each member being based on an individual member performance cost, a flexible task violation cost, and a behavior cost, the behavior cost being based on the initial value for the at least one of the selflessness trait and the selfishness trait;

receiving, by the calendar server, feedback from each member, the feedback including a selection of a feasible schedule of the first feasible schedules as a preferred schedule;

determining, by the calendar server, an updated value for the at least one of the selflessness trait and the selfishness trait of each member, the selfishness trait being assumed when the feasible schedule that is selected by the member places a higher total schedule cost on other members of the plurality of members, and the selflessness trait being assumed when the feasible schedule that is selected by the member places a lower total schedule cost on the other members;

generating, by the calendar server, a collaborative schedule based on the feasible schedules selected by the plurality of members;

storing, by the calendar server, the updated value for the at least one of the selflessness trait and the selfishness trait of each member based on the received feedback; and

calculating, by the calendar server, second collaborative schedule information based on the updated value for at least one of the selflessness trait and the selfishness trait of at least one member, the second collaborative schedule information including second feasible schedules for a second task, event, or responsibility.

2. The method of claim 1, wherein:

the calculating the first collaborative schedule information includes computing the total schedule cost for each member and for each of the feasible schedules, and

the total schedule cost for each member is computed according to total schedule cost expressions;

C_t(S_i,u_j)=C₁(S_i,u_j)−

C₂(S_i,u_j),

C₁(S_i,u_j)=C_p(S_i,u_j)−

γ

C_v(S_i,u_j); and

C₂(S_i,u_j)=Σ

_kα

_(j,k)max[C₁(S_i,u_j)−

C₁(S_i,u_k),0]+Σ

_kβ

_(j,k)max[C₁(S_i,u_k)−

C₁(S_i,u_j),0];

in which;

i, j, and k represent indexing variables;

S_i represents one of the feasible schedules identified by an indexing variable;

u_j represents one of the members identified by another indexing variable;

C_t(S_i,u_j) represents a total schedule cost for one of the members and for one of the feasible schedules;

C₁(S_i,u_j) represents individual costs incurred by one of the members and for one of the feasible schedules;

C₂(S_i,u_j) represents behavioral costs incurred by one of the members and for one of the feasible schedules;

C_p(S_iu_j) represents an individual member performance cost of one of the feasible schedules for one of the members;

C_v(S_i,u_j) represents a flexible task violation cost while performing one of the feasible schedules for one of the members;

γ

represents a penalty factor which decreases the individual costs if any feasible tasks are violated;

max represents a maximizing function;

α

_(j,k) represents a selflessness parameter that is independent of the feasible schedule;

β

_(j,k) represents a selfishness parameter that is independent of the feasible schedule;

Σ

_kα

_(j,k)max[C₁(S_i,u_j)−

C₁(S_i,u_k), 0] represents selflessness expression; and

Σ

_kβ

_(j,k)max[C₁(S_i,u_k)−

C₁(S_i,u_j), 0] represents selfishness expression.

3. The method of claim 2, wherein:

the determining the updated value for the at least one of the selflessness trait and the selfishness trait includes estimating a selflessness parameter and a selfishness parameter based on the received feedback; and

the estimating includes;

calculating a probability of each member responding to a neutral mutual task;

calculating a probability of each member responding to a constrained mutual task in a selfless manner; and

calculating a probability of each member responding to a constrained mutual task in a selfish manner.

4. The method of claim 3, wherein:

the calculating the probability of each member responding to the neutral mutual task is performed according to a neutral mutual task expression;

5. The method of claim 4, wherein the calculating the probability of each member responding to the constrained mutual task in the selfish manner and calculating the probability of each member responding to the constrained mutual task in the selfless manner includes:

plotting a histogram of similar situations; and

using Laplace smoothing, computing the selflessness parameter according to a selflessness expression;

α

_(i,j)=(σ

_i+1)/(N+s);

β

_(i,j)=(T_i+1)/(N+s); and

s=t COMB x;

in which;

s is a total number of assumable combinations in the constrained task;

t represents the traits;

x represents a number of members in the plurality of members; and

COMB represents a combination operator.

6. The method of claim 1, further comprising:

creating a table that includes the total schedule costs for each of the feasible schedules for each member; and

communicating the table to member devices associated with each member.

7. The method of claim 1, wherein:

the feedback includes a first selection in which a first feasible schedule is selected as the preferred schedule from a first member of the plurality of members, a second preference selecting the first feasible schedule from a second member, and the first feasible schedule having a higher total schedule cost to the first member than to the second member of the plurality of members; and

the method further comprises;

in response to a first preference, characterizing the first member as selfless; and

in response to the second preference, characterizing the second member as selfish.

8. The method of claim 1, wherein the feedback includes a rank of each of the feasible schedules relative to one another.

9. The method of claim 1, wherein the calculating the first collaborative schedule information includes:

enumerating all possible schedules that include completion of the individual member tasks and the mutual member tasks;

defining a vector for each of the individual member tasks and for each of the mutual member tasks, the vectors formatted as;

[start time, end time, start location, end location];

dividing the individual member tasks and the mutual member tasks into a particular number of task groups, wherein each of the task groups have the start time or the end time within a particular time frame;

for the individual member tasks and the mutual member tasks in each of the task groups, considering one or more possibilities for unknown location parameters;

choosing vectors that exhibit continuity in one or more unknown locations;

finding unknown values for the start times or for the end times that satisfy the individual member constraints and the mutual member constraints; and

identifying the feasible schedules as a subset of the possible schedules that are achievable without violating the individual member constraints or mutual member constraints.

10. The method of claim 1, wherein the individual member tasks and the mutual member tasks include at least one hard task that is characterized as not being subject to violation and at least one flexible task that is characterized as subject to violation.

11. A non-transitory computer-readable medium having encoded therein programming code executable by a processor to perform or control performance of operations comprising:

setting, by a calendar server, an initial value for at least one of a selflessness trait rL a selfishness trait of each member of a plurality of members who are included in a group to which a collaborative schedule pertains, the at least one of the selflessness trait or the selfishness trait related to behavior of a given member as indicated by preferences of the given member;

receiving, at the calendar server, group inputs that include individual member tasks, mutual member tasks, individual member constraints, and mutual member constraints;

receiving, at the calendar server, environmental data collected by a data collection engine;

calculating, by the calendar server, first collaborative schedule information based on the environmental data, the initial value for the at least one of the selflessness trait and the selfishness trait, and the group inputs, the first collaborative schedule information including a number of first feasible schedules for a first task, event, or responsibility and total schedule costs that are associated with the first feasible schedules for each member, the total schedule costs for each member being based on an individual member performance cost, a flexible task violation cost, and a behavior cost, the behavior cost being based on the initial value for the at least one of the selflessness trait and the selfishness trait;

receiving, by the calendar server, feedback from each member, the feedback including a selection of a feasible schedule of the first feasible schedules as a preferred schedule;

determining, by the calendar server, an updated value for the at least one of the selflessness trait and the selfishness trait of each member, the selfishness trait being assumed when the feasible schedule that is selected by the member places a higher total schedule cost on other members of the plurality of members, and the selflessness trait being assumed when the feasible schedule that is selected by the member places a lower total schedule cost on the other members;

generating, by the calendar server, a collaborative schedule based on the feasible schedules selected by the plurality of members;

storing, by the calendar server, the updated value for the at least one of the selflessness trait and the selfishness trait of each member based on the received feedback; and

calculating, by the calendar server, second collaborative schedule information based on the updated value for at least one of the selflessness trait and the selfishness trait of at least one member, the second collaborative schedule information including second feasible schedules for a second task, event, or responsibility.

12. The non-transitory computer-readable medium of claim 11, wherein:

the calculating the collaborative schedule information includes computing the total schedule cost for each member and for each of the feasible schedules, and

the total schedule cost for each member is computed according to total schedule cost expressions;

C_t(S_i,u_j)=C₁(S_i,u_j)−

C₂(S_i,u_j),

C₁(S_i,u_j)=C_p(S_i,u_j)−

γ

C_v(S_i,u_j); and

C₂(S_i,u_j)=Σ

_kα

_(j,k)max[C₁(S_i,u_j)−

C₁(S_i,u_k),0]+Σ

_kβ

_(j,k)max[C₁(S_i,u_k)−

C₁(S_i,u_j),0];

in which;

i, j, and k represent indexing variables;

S_i represents one of the feasible schedules identified by an indexing variable;

u_j represents one of the members identified by another indexing variable;

C_t(S_i,u_j) represents a total schedule cost for one of the members and for one of the feasible schedules;

C₁(S_i,u_j) represents individual costs incurred by one of the members and for one of the feasible schedules;

C₂(S_i,u_j) represents behavioral costs incurred by one of the members and for one of the feasible schedules;

C_p(S_i,u_j) represents an individual member performance cost of one of the feasible schedules for one of the members;

C_v(S_i,u_j) represents a flexible task violation cost while performing one of the feasible schedules for one of the members;

γ

represents a penalty factor which decreases the individual costs if any feasible tasks are violated;

max represents a maximizing function;

α

_(j,k) represents a selflessness parameter that is independent of the feasible schedule;

β

_(j,k) represents a selfishness parameter that is independent of the feasible schedule;

Σ

_kα

_(j,k)max[C₁(S_i,u_j)−

C₁(S_i,u_k), 0] represents selflessness expression; and

Σ

_kβ

_(j,k)max[C₁(S_i,u_k)−

C₁(S_i,u_j), 0] represents selfishness expression.

13. The non-transitory computer-readable medium of claim 12, wherein:

the determining the updated value for the at least one of the selflessness trait and the selfishness trait includes estimating a selflessness parameter and a selfishness parameter based on the received feedback; and

the estimating includes;

calculating a probability of each member responding to a neutral mutual task;

calculating a probability of each member responding to a constrained mutual task in a selfless manner; and

calculating a probability of each member responding to a constrained mutual task in a selfish manner.

14. The non-transitory computer-readable medium of claim 13, wherein:

the calculating the probability of each member responding to the neutral mutual task is performed according to a neutral mutual task expression;

15. The non-transitory computer-readable medium of claim 14, wherein the calculating the probability of each member responding to the constrained mutual task in the selfish manner and calculating the probability of each member responding to the constrained mutual task in the selfless manner includes:

plotting a histogram of similar situations; and

using Laplace smoothing, computing the selflessness parameter according to a selflessness expression;

α

_(i,j)=(σ

_i+1)/(N+s);

β

_(i,j)=(T_i+1)/(N+s); and

s=t COMB x;

in which;

s is a total number of assumable combinations in the constrained task;

t represents the traits;

x represents a number of members in the plurality of members; and

COMB represents a combination operator.

16. The non-transitory computer-readable medium of claim 11, wherein the operations further comprise:

creating a table that includes the total schedule costs for each of the feasible schedules for each member; and

communicating the table to member devices associated with each member.

17. The non-transitory computer-readable medium of claim 11, wherein:

the feedback includes a first selection in which a first feasible schedule is selected as the preferred schedule from a first member of the plurality of members, a second preference selecting the first feasible schedule from a second member, and the first feasible schedule having a higher total schedule cost to the first member than to the second member of the plurality of members; and

the operations further comprise;

in response to the first preference, characterizing the first member as selfless; and

in response to the second preference, characterizing the second member as selfish.

18. The non-transitory computer-readable medium of claim 11, wherein the feedback includes a rank of each of the feasible schedules relative to one another.

19. The non-transitory computer-readable medium of claim 11, wherein the calculating the collaborative schedule information includes:

enumerating all possible schedules that include completion of the individual member tasks and the mutual member tasks;

defining a vector for each of the individual member tasks and for each of the mutual member tasks, the vectors formatted as;

[start time, end time, start location, end location];

dividing the individual member tasks and the mutual member tasks into a particular number of task groups, wherein each of the task groups have the start time or the end time within a particular time frame;

for the individual member tasks and the mutual member tasks in each of the task groups, considering one or more possibilities for unknown location parameters;

choosing vectors that exhibit continuity in one or more unknown locations;

finding unknown values for the start times or for the end times that satisfy the individual member constraints and the mutual member constraints; and

identifying the feasible schedules as a subset of the possible schedules that are achievable without violating the individual member constraints or mutual member constraints.

20. The non-transitory computer-readable medium of claim 11, wherein the individual member tasks and the mutual member tasks include at least one hard task that is characterized as not being subject to violation and at least one flexible task that is characterized as subject to violation.