Integrated marketing and operations decisions-making under multi-brand competition

• US 6,009,407 A
• Filed: 02/27/1998
• Issued: 12/28/1999
• Est. Priority Date: 02/27/1998
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Garg, Amit
• Primary Examiner(s): Trammell, James P.
• Assistant Examiner(s): Nguyen, Cuong H.
• Application Number:
  US09/032,527
• Time in Patent Office: 669 Days
• Field of Search: 705/10, 705/14, 705/28, 705/7
• US Class Current: 705/7.25
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 10/0639 Performance analysis of employees; Performance analysis of enterprise or organisation operations
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0203 Market surveys; Market polls
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions

Abstract

A computer-implemented method for merging product marketing control and product inventory control, generates a segment-level consumer choice model for a plurality of competing brands, and aggregates that to a market-level consumer choice model, then generates a brand-level demand probability distribution function based on the choice models. A cost-minimized base stock level and a demand forecast for each of the plurality of brands is generated based on the market level model consumer choice model and on pricing, promotion, and other marketing data for each of the brands. An inventory control receives inventory subtraction data and inventory addition data and, using the cost-minimized base stock levels, generates orders to replenish the inventory.

Claims

1. A computer-implemented method for merged marketing management and inventory management, said method comprising the steps of:

(a) inputting a consumer preference data representing consumer survey information as to a preference toward each of a plurality of brands of competing goods, by each of a plurality of consumer segments;

(b) inputting a plurality of pricing information representing a corresponding price for each of said plurality of brands;

(c) inputting a plurality of promotions data characterizing marketing promotions relating to said plurality of brands;

(d) producing a segment-level consumer choice model relating said plurality of competing brands and said plurality of consumer segments, comprising sub-steps of;

(1) inputting a plurality of data λ

_h, for h=1 to H, each representing a total amount of purchases over a time period by consumers within a corresponding hth segment of a plurality of H consumer segments;

(2) calculating a plurality of data, A_ih, for i=1 to B and for h=1 to H, each representing an attraction of each of a plurality of B brands to each of said plurality of H consumer segments, said calculating based in part on said consumer preference data;

(3) calculating a plurality of relative market share data, S_ih, for i=1 to B and for h=1 to H, each representing a relative market share of each of said plurality of B brands relative to the remainder of said plurality of B brands in each of said plurality of H consumer segments, said calculating based on at least one of said data A_ih ; and

(4) generating a plurality of mean purchase data, λ

_ih, for i=1 to B and for h=1 to H, each of said data representing a mean of purchases of each of said plurality of B brands by consumers in each of said plurality of H segments, said generating based on said data λ

_h ;

##EQU19## where;

∝

_i is an attraction of brand i to said plurality of consumer segments, ∝

_i being based on said consumer preference data,

ε

_ih is a predetermine error term,

X_kj is a kth of a plurality of marketing mix variables for brand j, said plurality based on at least one of said plurality of pricing information data and said plurality of promotions data, ##EQU20## is a parameter to be estimated, and f_k (.) Is a predetermined monotone transformation of the marketing mix variable X_kj,

then, a relative market share data, ##EQU21## representing a relative market share of brands B relative to the remainder of brands B in each consumer segment H,

(e) generating a brand-level demand forecast data based on said segment-level consumer choice model, each of said output data being a demand forecast for a particular brand from among said plurality of brands in relation to a particular consumer segment from among said plurality of segments, said generating further based, in part, on said pricing data and said promotions data;

(f) generating a market-level consumer choice model by aggregating the segment-level model generated at step (d) over the plurality of segments;

(g) inputting a total period sales data representing, for each of said plurality of brands, a total sales over a given market over a given period of time;

(h) generating a market share data representing, for each of said plurality of brands, a market share in relation to the remainder of said brands, based on said total period sales data;

(i) characterizing a brand-level distribution of demand for each of said plurality of brands, said characterizing comprising sub-steps of;

(1) calculating a mean of a market share of each said plurality of brands, based on said relative market share data,

(2) calculating a joint probability distribution of demands for said plurality of brands, conditional on said total period sales data,

(3) calculating a marginal distribution of each of said plurality of brands, based on said mean calculated by step (i)(1) and said total period sales;

(j) inputting a plurality of brand unit cost data; and

(k) generating a plurality of inventory base stock level data, said data representing an optimal base stock level for a store inventory of each brand in the market, said generating based on said characterizing a brand-level distribution for each of said plurality of brands and on said brand unit cost data.

2. A computer-implemented method for merged marketing management and inventory management according to claim 1, further comprising steps of:

(1) receiving an inventory subtraction data representing a removal of units from an inventory of said plurality of brands;

(m) receiving an inventory addition data representing an addition of units to said inventory;

(n) generating an order data representing an order for a quantity of units of said plurality of brands for maintaining an inventory level in said inventory of at least one of said plurality brands, said generating based on said inventory base stock level data, said inventory subtraction data, and said inventory addition data.

3. A computer-implemented method for merged marketing management and inventory management according to claim 1, further comprising the step of generating a policy performance data representing performance value of the plurality of inventory base stock level data based on a plurality of predetermined metrics.

4. A computer-implemented method for merged marketing management and inventory management according to claim 1, further comprising the step of displaying said brand-level demand forecast data.

5. A computer-implemented method for merged marketing management and inventory management according to claim 1, wherein the sub-step (1) of the step (d) of producing a segment-level consumer choice model calculates said plurality of attraction data, A_ih, for i=1 to B and for h=1 to H, in accordance with:

##EQU22## where α

_i is an attraction of brand i to said plurality of consumer segments, α

_i being based on said consumer preference data,

ε

_ih is a predetermined error term,

X_kj is a kth of a plurality of marketing mix variables for brand j, said plurality based on at least one of said plurality of pricing information data and said plurality of promotions data,

β

_kij.sup.(h) is a parameter to be estimated, and

f_k (.) is a predetermined monotone transformation of the marketing mix variable, X_kj, and

wherein the sub-step (2) of the step (d) of calculating a segment-level consumer choice model calculates said relative market share data, S_ih, for i=1 to B and for h=1 to H, in accordance with;

##EQU23##

6. A computer-implemented method for merged marketing management and inventory management according to claim 1, wherein step (e) of generating a brand-level demand forecast data based on said segment-level consumer choice model comprises the sub-step of generating a market share elasticity data in accordance with:

where

MNL means Multi-nominal Logit,

MCI means Multiplicative Competitive Interaction variable,

X_kj is a kth of a plurality of marketing mix variables for brand j, said plurality based on at least one of said plurality of pricing information data and said plurality of promotions data, and

β

_kij.sup.(h) is a parameter to be estimated.

7. A computer-implemented method for merged marketing management and inventory management according to claim 1, wherein step (e) of generating a brand-level demand forecast data based on said segment-level consumer choice model comprises the sub-step of generating a market share elasticity data in accordance with:

##EQU24## where MNL means Multi-nominal Logit, MCI means Multiplicative Competitive Interaction variable,

X_kj is a kth of a plurality of marketing mix variables for brand j, said plurality based on at least one of said plurality of pricing information data and said plurality of promotions data, and

β

_kij.sup.(h) is a parameter to be estimated.

8. A computer-implemented method for merged marketing management and inventory management according to claim 6, wherein the step (e) of generating a brand-level demand forecast data based on said segment-level consumer choice model further comprises the sub-step of generating a total market share elasticity data in accordance with:

##EQU25## where q_h is an expected relative volume of segment h, calculated according to;

##EQU26##

9. A computer-implemented method for merged marketing management and inventory management according to claim 1, wherein the step (k) of generating a plurality of inventory base stock level data generates said data based on a non-linear program P, in accordance with:

subject to

space="preserve" listing-type="equation">P(S.sub.i ≧

Z.sub.i =1, . . . , B|m)≧

α

where h_i is a predetermined unit holding cost per period for items of brand i, I_i (S_i) is a pre-calculated average on-hand inventory of brand i, and α

is a fill-rate requirement for the brands i.

10. A computer-implemented method for merged marketing management and inventory management according to claim 1, wherein the step (k) of generating a plurality of inventory base stock level data generates said data based on a non-linear program P'"'"', in

accordance with:

##EQU27## subject to

space="preserve" listing-type="equation">P(S.sub.i ≧

Z.sub.i =1, . . . , B|m)≧

α

where h_i is a predetermined unit holding cost per period for items of brand i and S_i is the base stock level for brand i, and α

is a fill-rate requirement for the brands i.