Interactive method to reduce the amount of tradeoff information required from decision makers in multi-attribute decision making under uncertainty

• US 9,251,461 B2
• Filed: 03/15/2013
• Issued: 02/02/2016
• Est. Priority Date: 03/15/2013
• Status: Expired due to Fees

• Current Assignee: University College Cork (Higher Education Authority), International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Abdul, Razak, Deleris, Lea A., Marinescu, Radu, Wilson, Peter Nicholas
• Primary Examiner(s): CHANG, LI WU
• Application Number:
  US13/839,211
• Publication Number:
  US 20140279801A1
• Time in Patent Office: 1,054 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06N 5/00 Computing arrangements using knowledge-based models
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 5/04 Inference or reasoning models

Abstract

There is provided a method, a system and a computer program product for supporting a decision making process. The system receives a decision model from a decision maker, the decision model used for determining a solution to a decision problem based on attributes and uncertainties of the decision problem. The decision problem includes information about a plurality of outcome vectors that represent all possible outcomes and the uncertainties associated with the decision problem. The system determines whether the received decision model can be solved without receiving any preference information from the decision maker. The system receives partially specified preference information from the decision maker if the received decision model cannot be solved without any preference information. The system solves the decision model with the partially specified preference information. The system recommends, based on the solution, one or more decisions to the decision maker.

Claims

1. A method for supporting a decision making process, the method comprising:

implementing a decision model for determining a solution to a decision problem based on attributes and uncertainties of the decision problem, the decision problem including information about a plurality of outcome vectors that represent all possible outcomes and the uncertainties associated with the decision problem, the decision model precluding any preference information of the attributes and the outcome vectors;

determining whether the decision model can be solved without receiving any preference information from a decision maker;

receiving partially specified preference information from the decision maker if the received decision model cannot be solved without any preference information, said receiving partially specified preference information comprising;

identifying N pairs of outcome vectors to present to the decision maker for preference assessment, said N pairs of outcome vectors identified by;

selecting the N pair of outcome vectors that maximizes an expected information gain, the expected information gain including;

a reduction in a number of preferred strategies;

or

selecting the N pair of outcome vectors that balances user-friendliness and the expected information gain;

presenting N elicitation questions, based on the identified N pairs of outcome vectors, to the decision maker, an elicitation question requesting of the decision maker which outcome vector the decision maker prefers between two outcome vectors; and

providing, by the decision maker, answers to the N elicitation questions as representing the partially specified preference information;

solving the decision model with the partially specified preference information; and

recommending, based on the solution, one or more decisions to the decision maker,

wherein a processor coupled to a memory device perform steps of;

receiving the decision model, determining, receiving the partially specified preference information, solving, and recommending.

2. The method according to claim 1, wherein the step of determining includes steps of:

identifying a plurality of strategies integrated into the decision model; and

eliminating one or more strategies, among the identified plurality of strategies.

3. The method according to claim 1, further comprising:

setting L to be a set of input outcome vectors of the decision problem;

setting a first set that stores N pairs of outcome vectors to be empty;

enumerating all pairs of outcome vectors (u, v) in the set L;

if the decision maker prefers an outcome vector u over an outcome vector v, setting a score of a pair of outcome vectors (u, v) to a number of strategies obtained by solving the decision problem under an assumption that the outcome vector u is preferred to the outcome vector v;

if the decision maker prefers an outcome vector v over an outcome vector u, setting a score of a pair of outcome vectors (v, u) to a number of strategies obtained by solving the decision problem under an additional assumption that the outcome vector v is preferred to the outcome vector u;

sorting all the pairs of outcome vectors according an ascending order of corresponding scores;

storing the sorted pairs of outcome vectors in a third set;

for each pair of outcome vector (u, v) in the third set, if the decision confirms that the outcome vector u is preferred over the outcome vector v, adding the pair of outcome vector (u, v) to the first set;

solving the decision problem with the first set;

evaluating whether the solution satisfies the decision maker;

if the decision maker does not satisfy the solution, generating outcome vectors u and v on a boundary of the first set;

if the decision maker confirms that the outcome vector u is preferred over the outcome vector v, adding the pair of outcome vector (u, v) to the first set; and

repeating the solving, the evaluating, the generating, and the adding the pair of outcome vector (u, v) to the first set until the solution satisfies the decision maker.

4. The method according to claim 1, further comprising:

repeating the receiving the partially specified preference information and the solving the decision model until a number of the recommended decisions satisfies a rule or is chosen by the decision maker.

5. The method according to claim 1, wherein the partially specified preference information is a minimum amount of preference information.

6. A system for providing a decision support, the system comprising:

a memory device;

a processor coupled to the memory device, wherein the processor is configured to perform steps of;

implementing a decision model for determining a solution to a decision problem based on attributes and uncertainties of the decision problem, the decision problem including information about a plurality of outcome vectors that represent all possible outcomes and the uncertainties associated with the decision problem, the received decision model precluding any preference information of the attributes and the outcome vectors;

determining whether the decision model can be solved without receiving any preference information from a decision maker;

receiving partially specified preference information from the decision maker if the received decision model cannot be solved without any preference information, said receiving partially specified preference information comprising;

identifying N pairs of outcome vectors to present to the decision maker for preference assessment, said N pairs of outcome vectors identified by;

selecting the N pair of outcome vectors that maximizes an expected information gain, the expected information gain including;

a reduction in a number of preferred strategies;

or

selecting the N pair of outcome vectors that balances user-friendliness and the expected information gain;

presenting N elicitation questions, based on the identified N pairs of outcome vectors, to the decision maker, an elicitation question requesting of the decision maker which outcome vector the decision maker prefers between two outcome vectors; and

providing, by the decision maker, answers to the N elicitation questions as representing the partially specified preference information;

solving the decision model with the partially specified preference information; and

recommending, based on the solution, one or more decisions to the decision maker.

7. The system according to claim 6, wherein the processor is further configured to perform steps of:

identifying a plurality of strategies integrated into the decision model; and

eliminating one or more strategies, among the identified plurality of strategies.

8. The system according to claim 6, wherein the processor is further configured to perform steps of:

setting L to be a set of input outcome vectors of the decision problem;

setting a first set that stores N pairs of outcome vectors to be empty;

enumerating all pairs of outcome vectors (u, v) in the set L;

if the decision maker prefers an outcome vector u over an outcome vector v, setting a score of a pair of outcome vectors (u, v) to a number of strategies which obtained by solving the decision problem under an assumption that the outcome vector u is preferred to the outcome vector v;

if the decision maker prefers an outcome vector v over an outcome vector u, setting a score of a pair of outcome vectors (v, u) to a number of strategies obtained by solving the decision problem under an additional assumption that the outcome vector v is preferred to the outcome vector u;

sorting all the pairs of outcome vectors according an ascending order of corresponding scores;

storing the sorted pairs of outcome vectors in a third set;

for each pair of outcome vector (u, v) in the third set, if the decision confirms that the outcome vector u is preferred over the outcome vector v, adding the pair of outcome vector (u, v) to the first set;

solving the decision problem with the first set;

evaluating whether the solution satisfies the decision maker;

if the decision maker does not satisfy the solution, generating outcome vectors u and v on a boundary of the first set;

if the decision maker confirms that the outcome vector u is preferred over the outcome vector v, adding the pair of outcome vector (u, v) to the first set; and

repeating the solving, the evaluating, the generating, and the adding the pair of outcome vector (u, v) to the first set until the solution satisfies the decision maker.

9. The system according to claim 6, wherein the processor is further configured to perform a step of:

repeating the receiving the partially specified preference information and the solving the decision model until a number of the recommended decisions satisfies a rule or is chosen by the decision maker.

10. A computer program product for providing a decision support, the computer program product comprising a non-transitory storage medium readable by a processing circuit and storing instructions run by the processing circuit for performing a method, said method steps comprising:

implementing a decision model for determining a solution to a decision problem based on attributes and uncertainties of the decision problem, the decision problem including information about a plurality of outcome vectors that represent all possible outcomes and the uncertainties associated with the decision problem, the received decision model precluding any preference information of the attributes and the outcome vectors;

determining whether the decision model can be solved without receiving any preference information from a decision maker;

receiving partially specified preference information from the decision maker if the received decision model cannot be solved without any preference information, said receiving partially specified preference information comprising;

identifying N pairs of outcome vectors to present to the decision maker for preference assessment, said N pairs of outcome vectors identified by;

selecting the N pair of outcome vectors that maximizes an expected information gain, the expected information gain including;

a reduction in a number of preferred strategies;

or

selecting the N pair of outcome vectors that balances user-friendliness and the expected information gain;

presenting N elicitation questions, based on the identified N pairs of outcome vectors, to the decision maker, an elicitation question requesting of the decision maker which outcome vector the decision maker prefers between two outcome vectors; and

providing, by the decision maker, answers to the N elicitation questions as representing the partially specified preference information;

solving the decision model with the partially specified preference information; and

recommending, based on the solution, one or more decisions to the decision maker.

11. The computer program product according to claim 10, further comprising:

repeating the receiving the partially specified preference information and the solving the decision model until a number of the recommended decisions satisfies a rule or is chosen by the decision maker.

12. The computer program product according to claim 10, wherein the partially specified preference information is a minimum amount of preference information.