Methods for parts procurement quantity determination where demand is uncertain for the product in which the parts are used
First Claim
Patent Images
1. A stochastic method for determining, over a series of time periods, parts procurement requirements for products, certain said parts used in a plurality of products, wherein actual demand for said products is unknown, said method adapted to meet a specified service level for all products and to minimize expected excess part inventories, the method having inputs comprising, lists of parts for each said product, prices for said parts, and forecasts of demands for each said product in each time period, the method comprising computer implemented steps of:
- a. deriving a first cumulative demand D'"'"'ci,t for each part ci through a current time period t, and a second cumulative demand Dci,t for each part ci through a time period t-1, added to the demand for each part in current time period t, modified by a specified volume fraction of actual demand for each part in period t that is to be satisfied in period t, each cumulative demand expressed by a mean and standard deviation;
b. determining a minimum cumulative procurement quantity q'"'"'*ci,t for each part through time period t that meets said specified service level, from a demand distribution function FDci,t from said second cumulative demand;
c. finding a factor (λ
), for each part, at said determined minimum cumulative procurement quantity, that satisfies equation 6t ;
##EQU14## where;
ci is part i,Pci is the cost of part i,t is the time period,q'"'"'ci,t is the cumulative procurement of part i up to period t,FD'"'"'ci,t (q'"'"'ci,t) is the cumulative distribution function of the random variable q'"'"'ci,t,* means, at optimality,λ
t is the LaGrange Multiplier for time period t,ƒ
Dci,t (q'"'"'ci,t) is the density function of the random variable q'"'"'ci,t,FDci,t (q'"'"'ci,t) is the cumulative distribution function of the random variable q'"'"'ci,t, i means for all i, i.e., i=1,2, . . .d. employing a maximum value factor (λ
) found in step c to derive for each part, a new cumulative procurement quantity from equation 6t over all time periods through t, and based thereon, an overall service level for each part, said overall service level for each part determined by taking the product of cumulative demand distribution functions for all parts, as determined from said second cumulative demands, evaluated at said new cumulative procurement quantity;
e. finding if said new cumulative procurement quantities for all parts provide a service level equal to said specified service level, and, if not, revising said factor (λ
), and repeating steps d and e employing the revised factor (λ
) until said specified service level is achieved; and
f. employing resultant procurement quantities that enable achievement of the service level for acquisition of said parts.
1 Assignment
0 Petitions
Accused Products
Abstract
Methods are described for predicting parts procurement requirements for products over a plurality of time periods, with certain of the parts being common to a plurality of products. The actual demand for the products is unknown, but the method assures that a specified service level is met for all products and minimizes expected excess part inventories. The methods are provided with inputs which, among others, includes lists of parts for each product, prices for the parts, and demand forecasts for each product in each time period, each forecast in the form of a mean and standard deviation. The description of the problem includes an objective function of minimizing expected excess inventory while satisfying the constraint that a specified service level be achieved. The problem is transformed into an unconstrained problem through the use of a Lagrange multiplier. The solution is achieved by performing a one parametric search on the value of the multiplier. The solution may achieve higher service levels than specified. Additional methods are described for improving the procurement decisions to more closely meet the service requirement.
-
Citations
7 Claims
-
1. A stochastic method for determining, over a series of time periods, parts procurement requirements for products, certain said parts used in a plurality of products, wherein actual demand for said products is unknown, said method adapted to meet a specified service level for all products and to minimize expected excess part inventories, the method having inputs comprising, lists of parts for each said product, prices for said parts, and forecasts of demands for each said product in each time period, the method comprising computer implemented steps of:
-
a. deriving a first cumulative demand D'"'"'ci,t for each part ci through a current time period t, and a second cumulative demand Dci,t for each part ci through a time period t-1, added to the demand for each part in current time period t, modified by a specified volume fraction of actual demand for each part in period t that is to be satisfied in period t, each cumulative demand expressed by a mean and standard deviation; b. determining a minimum cumulative procurement quantity q'"'"'*ci,t for each part through time period t that meets said specified service level, from a demand distribution function FDci,t from said second cumulative demand; c. finding a factor (λ
), for each part, at said determined minimum cumulative procurement quantity, that satisfies equation 6t ;
##EQU14## where;
ci is part i,Pci is the cost of part i, t is the time period, q'"'"'ci,t is the cumulative procurement of part i up to period t, FD'"'"'ci,t (q'"'"'ci,t) is the cumulative distribution function of the random variable q'"'"'ci,t, * means, at optimality, λ
t is the LaGrange Multiplier for time period t,ƒ
Dci,t (q'"'"'ci,t) is the density function of the random variable q'"'"'ci,t,FDci,t (q'"'"'ci,t) is the cumulative distribution function of the random variable q'"'"'ci,t, i means for all i, i.e., i=1,2, . . . d. employing a maximum value factor (λ
) found in step c to derive for each part, a new cumulative procurement quantity from equation 6t over all time periods through t, and based thereon, an overall service level for each part, said overall service level for each part determined by taking the product of cumulative demand distribution functions for all parts, as determined from said second cumulative demands, evaluated at said new cumulative procurement quantity;e. finding if said new cumulative procurement quantities for all parts provide a service level equal to said specified service level, and, if not, revising said factor (λ
), and repeating steps d and e employing the revised factor (λ
) until said specified service level is achieved; andf. employing resultant procurement quantities that enable achievement of the service level for acquisition of said parts. - View Dependent Claims (2, 3, 4, 5)
-
-
6. A stochastic method for predicting, over a series of time periods, parts procurement requirements for products, certain said parts used in a plurality of products, wherein actual demand for said products is unknown, said method adapted to meet a specified service level for all products and to minimize expected excess part inventories, the method employing lists of parts for each said product, prices for said parts, and probability forecasts of demands for each said product in each time period, the method comprising the machine implemented steps of:
-
a. for each part, determining a random demand value from product bills of materials and product random demand values; b. finding a value of λ
from equation 6t by substituting therein, said part random demand value in place of q'"'"'ci,t. ##EQU15## where;
ci is part i,Pci is the cost of part i, t is the time period, q'"'"'ci,t is the cumulative procurement of part i up to period t, FD'"'"'ci,t (q'"'"'ci,t) is the cumulative distribution function of the random variable q'"'"'ci,t, * means, at optimality, λ
t is the LaGrange Multiplier for time period t,ƒ
Dci,t (q'"'"'ci,t) is the density function of the random variable q'"'"'ci,t,FDci,t (q'"'"'ci,t) is the cumulative distribution function of the random variable q'"'"'ci,t, i means for all i, i.e., i=1,2, . . . c. In all parts, finding λ
max values and placing said values in an ordered list;d. repeating steps a-c and after n iterations selecting a value of λ
max at a percentile position in said list that corresponds to said specified service value; ande. employing said selected λ
max and equation 6t to find a new value of q'"'"'ci,tf. employing the new cumulative procurement value q'"'"'ci,t to achieve said specified service level. - View Dependent Claims (7)
-
Specification
-
- Resources
Litigation Campaign Assessment
Thank you for your request. You will receive a custom alert email when the Litigation Campaign Assessment is available.
×
-
Current AssigneeInternational Business Machines Corporation
-
Original AssigneeInternational Business Machines Corporation
-
InventorsJayaraman, Rangarajan, Roundy, Robin, Srinivasan, Ramesh
-
Primary Examiner(s)Weinhardt, Robert A.
-
Application NumberUS07/698,632Time in Patent Office1,012 DaysField of Search364/400, 364/401, 364/402, 364/403, 364/468, 364/149, 364/152, 364/143US Class Current705/7.31CPC Class CodesG06Q 10/06 Resources, workflows, human...G06Q 10/087 Inventory or stock manageme...G06Q 30/0202 Market predictions or forec...
