COMPUTER IMPLEMENTED DECISION SUPPORT METHOD & SYSTEM

• US 20110270646A1
• Filed: 07/13/2009
• Published: 11/03/2011
• Est. Priority Date: 07/11/2008
• Status: Active Application

• Current Assignee: Abhilasha Aswal, Anushka Chandra Babu, Dileep Kumar, Gorur Narayana Srinivasa Prasanna, Mabel Mary Joy, Piyushkumar Jain
• Original Assignee: Abhilasha Aswal, Anushka Chandra Babu, Dileep Kumar, Gorur Narayana Srinivasa Prasanna, Mabel Mary Joy, Piyushkumar Jain
• Inventors: Prasanna, Gorur Narayana Srinivasa, Kumar, Dileep, Aswal, Abhilasha, Joy, Mabel Mary, Chandra Babu, Anushka, Jain, Piyushkumar
• Application Number:
  US13/003,507
• Publication Number:
  US 20110270646A1
• Time in Patent Office: Days
• US Class Current: 705/7.27
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/0633 Workflow analysis

Abstract

In this research, we propose to extend the robust optimization technique and target it for problems encountered in supply chain management. Our method represents uncertainty as polyhedral uncertainty sets made of simple linear constraints derivable from macroscopic economic data. We avoid the probability distribution estimation of stochastic programming. The constraints in our approach are intuitive and meaningful. This representation of uncertainty is applied to capacity planning and inventory optimization problems in supply chains. The representation of uncertainty is the unique feature that drives this research. It has led us to explore different problems in capacity/inventory planning under this new paradigm. A decision support system package has been developed, which can conveniently interface to manufacturing/firm data warehouses, inferring and analyzing constraints from historical data, analyzing performance (worst case/best case), and optimizing plans.

Claims

1-30. -30. (canceled)

31. A Computer implemented Decision Support method, comprising the step of feeding information in the form of at least one constraint set defined over a space of parameters, with a parameter being a multidimensional vector, with a constraint set having at least one constraint defined over said parameters, with allowable parameters satisfying all the constraints in at least one said constraint set, and offering facilities for at least one of:

a. determining at least one of set-theoretic relations, inclusive of subset, disjoint, and intersection, or at least one of metric relations, inclusive of maximum and minimum distances, between a first said constraint set and a said second constraint set, in an extended relational algebra engine;

b. transformation of a first said constraint set to obtain a second said constraint set having the same, greater, or smaller multidimensional volume using at least one of scaling, rotation, translations, and volume preserving, respectively volume increasing, respectively volume decreasing, general linear or non-linear transformations;

c. determining information content of a said constraint set by determining the volume of said constraint set in an information theory engine; and

d. and having a facility to determine a parameter, which satisfies all constraints in a first constraint set, and where a specified objective function defined over said parameters is maximized over all parameters satisfying all constraints in same said first constraint set.

32. The method of claim 31, where a first maximum and a first minimum, and a first difference between said first maximum and said first minimum, of the said objective function are determined over all parameters satisfying all constraints in said first constraint set.

33. The method of claim 32, where a second difference between a second maximum and a second minimum of said objective function, is determined over all parameters satisfying all constraints in a second said constraint set.

34. The method of claim 33, where volume and information content of first said constraint set, and volume and information content of second said constraint set is determined.

35. The method of claim 32, where said first constraint set is transformed using one of said transformation facilities to reduce the said difference.

36. The method of claim 31, with a said parameter, being a vector whose components are values of a set of variables in a supply chain management system, said values being either restricted to integers, or allowed to have real number values, and said variables representing one of (a) demand, (b) supply, (c) inventory, (d) cost, (e) revenue or (f) profit or other relevant variables of an entity in a supply chain management system.

37. The method of claim 36, where the value of a said variable is read from the database of said supply chain management system, said variable value or values being updated in realtime by input to said supply chain management system.

38. The method of claim 37, where said facility gives a signal indicating satisfaction or non-satisfaction of at least one of said constraint sets, or satisfaction or non-satisfaction of a complex query on said constraint sets, by said variable value or values.

39. The method of claim 36, where a constraint set is obtained from at least one of user input, prediction from data or constraints present in said supply chain management database, or transformation of a second constraint set present in said supply chain management database, where said prediction utilises an apriori constraint about the constraint set, and creates a constraint set which is a best approximation to the convex hull of the parameters in said supply chain management database, said apriori constraint being one of:

a. the value of a constraint coefficient is fixed;

b. the sum of all constraint coefficients is fixed;

c. the mean square sum of all constraint coefficients is fixed.

40. The method of claim 36, where said transformation is a volume preserving linear transformation or translation applied to said second constraint set.

41. The method of claim 36, where the constraint set obtained by transformation uses one of a general volume preserving nonlinear transformation, which may change the number of constraints in said constraint set or a non-volume preserving transformation.

42. The method of claim 36, where the constraint set obtained by transformation is stored back in a said supply chain management database,

43. The method of claim 36 where an optimal inventory policy is obtained either by using said constraint set in the input to a linear or convex or mixed integer linear or convex programming problem or by determining a trigger constraint set and a reorder constraint set, at least one of said trigger and reorder constraint sets involving more than one said supply chain variable, where said inventory policy initiates a supply chain reorder action, when a said supply chain variable or variables result in a parameter being included in the trigger constraint set, said supply chain reorder action moving said parameter to a point in the reorder constraint set.

44. The method of claim 36 where the results are analyzed in an output analyzer which offers facilities to look at the aggregates of said variables in a subset of the supply chain, said aggregrates being at least one of sum, maximum, or minimum, or other relevant analytics, said subset comprised of at least one of a supply chain node or edge.

45. A Computer implemented Decision Support system, comprising the means of feeding information in the form of at least one constraint set defined over a space of parameters, with a parameter being a multidimensional vector, with a constraint set having at least one constraint defined over said parameters, with allowable parameters satisfying all the constraints in at least one said constraint set, and means to invoke facilities for at least one of:

a. determining at least one of set-theoretic (subset, disjoint, and intersection) and metric relations, inclusive of distances between a first said constraint set and a said second constraint set, in an extended relational algebra engine;

b. transformation of a first said constraint set to obtain a second said constraint set having the same, greater, or smaller multidimensional volume using at least one of scaling, rotation, translations, and volume preserving, respectively volume increasing, respectively volume decreasing, general linear or non-linear transformations;

c. determining Information content of a said constraint set by determining the volume of said constraint set in an information theory engine;

d. and having a facility to determine a parameter, which satisfies all constraints in a first constraint set, and where a specified objective function defined over said parameters is maximized over all parameters satisfying all constraints in same said first constraint set.