System for creating optimized promotion event calendar
View Patent ↗Methods and apparatus for creating a promotional event calendar are provided. A sales model and a cost model are created The value of offers and promotional events using the sales model and cost model is determined. Combinations of offers and promotional events based on determined value are selected to create a promotion event calendar subject to the conditions from at least one store. The apparatus includes an econometric engine for modeling sales as a function of price to create a sales model, a financial model engine for modeling costs to create a cost model, a promotional engine coupled to the econometric engine, and financial model engine to receive input from the econometric engine and financial model engine. The promotional engine analyzes a plurality of offers and a plurality of promotional events to optimally match offers with promotional events to create a promotional event calendar.
1. A computerized price optimization system configured to create a promotional event calendar, in association with at least one store, the computerized price optimization system comprising:
an econometric engine, executed by a processor, configured to create a sales model, wherein the econometric engine includes an imputed variable generator for:
imputing base price variable and a promotional variable,
generating an equivalent price and an equivalent unit for each product of the plurality of demand groups using an equivalizing factor, and
wherein the sales model includes the imputed base price variable, the equivalent price, the equivalent unit and the promotional variable;
a financial model engine configured to create a cost model; and
a promotional engine coupled to the econometric engine and financial model engine, the promotional engine configured to:
receive input from the econometric engine and financial model engine;
receive manufacturer conditions from more than one manufacturer, wherein the manufacturer conditions are requirements the more than one manufacturer places on an offer, and the requirements include at least one promotional vehicle requirement;
receive store constraints from at least one store, and wherein the store constraints from the at least one store includes a linear constraint and a nonlinear constraint;
compute a score that links each product with one promotional vehicle type for each offer of a plurality of offers and for each promotional event of a plurality of promotional events, wherein the computing the score independently computes a value of each offer and a value of each promotional event using the sales model and cost model, and wherein each promotional event includes at least one type of promotional vehicle;
select combinations of offers from the plurality of offers, and promotional events from the plurality of promotional events to form a subset of offers and promotional events, wherein the selection of the combination of offers and events is based on the computed scores, the manufacturer conditions and store constraints;
reconcile the type of promotional vehicle associated with each promotional event in the subset with the promotional vehicle requirements for each offer in the subset; and
construct a promotion calendar utilizing the subset of offers and promotional events, and reconciled type of promotional vehicle and promotional vehicle requirements wherein the promotion calendar sets the promotion levels of each product.
2. The system, as recited in claim 1 , wherein the promotional engine further comprises a temporary price reduction optimizing engine configured to temporarily reduce prices after the promotional events and offers have been selected.
3. The system, as recited in claim 2 , further comprising a support tool connected to the promotional engine wherein the support tool is configured to receive the promotional event calendar from the promotional engine and provides a user interface to a client, wherein the user interface provides the promotional event calendar to the client.
4. The system as recited in claim 1 , wherein the constraints from the at least one store include at least one of ad space capacity and display space capacity.
5. The system as recited in claim 1 , wherein the constraints from the at least one store include at least one of an event type, a number of events, a brand promotion frequency, and a product promotion frequency.
6. The system as recited in claim 1 , wherein the selecting of offers with promotional events includes solving an integer problem using the linear constraint and the nonlinear constraint.
7. The system as recited in claim 1 , wherein the conditions from the at least one manufacturer include providing at least one of a promotional event and a specific amount of promotion.
8. The system as recited in claim 1 , wherein the conditions from the at least one manufacturer include not providing a promotional event for a competitor's product.
9. The system as recited in claim 1 , wherein the imputed base price variable is a function of initial base prices.
10. The system as recited in claim 9 , wherein the initial base prices are averages of non-promoted prices.
11. The system as recited in claim 1 , wherein the econometric engine utilizes Bayesian Shrinkage modeling which relies on a mixed-model methodology of combining data to dampen any extreme values of the data.
12. The system, as recited in claim 1 , wherein the more than one manufacturer includes a plurality of manufacturers and wholesalers.
13. A computer-implemented method for creating a promotional event calendar, in association with at least one store, the method for creating a promotional event calendar comprising:
creating a sales model, using a processor, for each product and wherein the sales model includes an imputed base price variable, an equivalent price, an equivalent unit and a promotional variable;
creating a cost model;
determining conditions from more than one manufacturer, wherein the conditions are requirements the more than one manufacturer places on an offer, and the requirements include at least one promotional vehicle requirement;
determining store constraints from at least one store, and wherein the store constraints from the at least one store includes a linear constraint and a nonlinear constraint;
computing a score that links each product with one promotional vehicle type for each offer of a plurality of offers and for each promotional event of a plurality of promotional events, wherein the computing the score independently computes a value of each offer and a value of each promotional event using the sales model and cost model, and wherein each promotional event includes at least one type of promotional vehicle;
selecting combinations of the offers and promotional events by optimizing the computed scores subject to the conditions from the more than one manufacturer and constraints from the at least one store;
reconciling the type of promotional vehicle associated with each promotional event in the subset with the promotional vehicle requirements for each offer in the subset; and
constructing a promotion calendar utilizing the subset of offers and promotional events, and reconciled type of promotional vehicle and promotional vehicle requirements wherein the promotion calendar sets the promotion levels of each product.
14. The computer-implemented method, as recited in claim 13 , further comprising the step of estimating net profit from the selected combination of offers and promotional events using the sales model and cost model.
15. The computer-implemented method as recited in claim 13 , wherein the constraints from the at least one store include at least one of ad space capacity and display space capacity.
16. The computer-implemented method as recited in claim 13 , wherein the constraints from the at least one store include at least one of an event type, a number of events, a brand promotion frequency, and a product promotion frequency.
17. The computer-implemented method as recited in claim 13 , wherein the selection of combinations of offers and promotional events includes solving an integer problem using the linear constraint and the nonlinear constraint.
18. The computer-implemented method as recited in claim 13 , wherein the conditions from the at least one manufacturer include providing at least one of a promotional event and a specific amount of promotion.
19. The computer-implemented method as recited in claim 13 , wherein the conditions from the at least one manufacturer include not providing a promotional event for a competitor's product.
20. The computer-implemented method as recited in claim 13 , wherein the more than one manufacturer includes a plurality of manufacturers and wholesalers.
21. A computer-implemented method for creating a promotional event calendar, comprising:
creating, using a processor, a sales model, comprising:
creating a plurality of demand groups, wherein each demand group is a set of at least one product and wherein at least one of the demand groups is a set of at least two products;
generating an equivalent price and an equivalent unit for each product of the plurality of demand groups using an equivalizing factor;
creating a sales model for each demand group and wherein the sales model for each demand group includes an imputed base price variable, the equivalent price, the equivalent unit and a promotional variable; and
creating a market share model for each product in each demand group;
creating a cost model;
determining conditions from more than manufacturer, wherein the conditions are requirements the more than one manufacturer places on an offer, and the requirements include at least one promotional vehicle requirement;
determining store constraints from at least one store, and wherein the store constraints from the at least one store includes a linear constraint and a nonlinear constraint;
computing a score that links each product with one promotional vehicle type for each offer of a plurality of offers and for each promotional event of a plurality of promotional events, wherein the computing the score independently computes a value of each offer and a value of each promotional event using the sales model and cost model, and wherein each promotional event includes at least one type of promotional vehicle;
selecting combinations of offers and promotional events to form a subset of offers and promotional events by optimizing the computed scores subject to the conditions from the more than one manufacturer and constraints from the at least one store;
reconciling the type of promotional vehicle associated with each promotional event in the subset with the promotional vehicle requirements for each offer in the subset;
estimating net profit from the selected combination of offers and promotional events using the sales model and cost model; and
construct a promotion calendar utilizing the subset of offers and promotional events, and reconciled type of promotional vehicle and promotional vehicle requirements wherein the promotion calendar sets the promotion levels of each product.
22. The computer-implemented method as recited in claim 21 , wherein the constraints from the at least one store include at least one of ad space capacity and display space capacity.
23. The computer-implemented method as recited in claim 21 , wherein the constraints from the at least one store include at least one of an event type, a number of events, a brand promotion frequency, and a product promotion frequency.
24. The computer-implemented method as recited in claim 21 , wherein the selection of combinations of offers and promotional events includes solving an integer problem using the linear constraint and the nonlinear constraint.
25. The computer-implemented method as recited in claim 21 , wherein the conditions from the at least one manufacturer include providing at least one of a promotional event and a specific amount of promotion.
26. The computer-implemented method as recited in claim 21 , wherein the conditions from the at least one manufacturer include not providing a promotional event for a competitor's product.
27. The computer-implemented method as recited in claim 21 , wherein the more than one manufacturer includes a plurality of manufacturers and wholesalers.
28. A computer-implemented method for creating a promotional event calendar, useful in association with at least one store, the computer-implemented method comprising:
generating an equivalent price and an equivalent unit for each product of a plurality of demand groups using an equivalizing factor;
creating, using a processor, a sales model for each demand group and wherein the sales model for each demand group includes an imputed base price variable, the equivalent price, the equivalent unit and a promotional variable;
modeling costs to create a cost model;
receiving manufacturer conditions from more than one manufacturer, wherein the manufacturer conditions are requirements the more than one manufacturer places on an offer, and the requirements include at least one promotional vehicle requirement;
receiving store constraints from at least one store, and wherein the store constraints from the at least one store includes a linear constraint and a nonlinear constraint;
computing a score that links each product with one promotional vehicle type for each offer of a plurality of offers and for each promotional event of a plurality of promotional events, wherein the computing the score independently computes a value of each offer and a value of each promotional event using the sales model and cost model, and wherein each promotional event includes at least one type of promotional vehicle;
selecting combinations of offers from the plurality of offers, and promotional events from the plurality of promotional events to form a subset of offers and promotional events, wherein the selection of the combination of offers and events is based on the computed scores, the manufacturer conditions and store constraints;
reconciling the type of promotional vehicle associated with each promotional event in the subset with the promotional vehicle requirements for each offer in the subset; and
constructing a promotion calendar utilizing the subset of offers and promotional events, and reconciled type of promotional vehicle and promotional vehicle requirements wherein the promotion calendar sets the promotion levels of each product.
29. The computer-implemented method, as recited in claim 28 , wherein the computing the score independently computes a value of each offer and a value of each event, and wherein the computing the score also links each product with one promotional vehicle type of the at least one promotional vehicle type.
30. The computer-implemented method, as recited in claim 28 , wherein selecting the combination of offers and events includes maximizing the sum of the values of the offers and events within the subset of offers and events, while adhering to the conditions from at least one manufacturer and the constraints from the at least one store.
31. The computer-implemented method, as recited in claim 28 , wherein reconciling the type of promotional vehicle with the promotional vehicle requirements includes solving an integer problem.
32. The computer-implemented method, as recited in claim 28 , wherein constructing a promotion calendar includes computing the resulting profit over a promotional period.
33. The computer-implemented method, as recited in claim 28 , wherein the selection subset of offers and events includes the integer programming model with the following objective function and appropriate restrictions on the variables:
∑
t
=
1
T
∑
o
Value
o
IO
o
,
t
+
∑
t
=
1
T
∑
e
Value
e
,
t
IE
e
,
t
where,
t: a time indices;
O: a set of offers;
o: an offer in O;
E: a set of events;
e: an event in E;
T: total planning period in weeks;
IE
e
,
t
≡
{
1
if
event
is
scheduled
to
start
in
period
t
0
otherwise
;
and
IO
o
,
t
≡
{
1
if
offer
o
is
accepted
in
period
t
0
otherwise
.
34. The computer-implemented method, as recited in claim 28 , wherein the reconciliation includes the integer programming model with the following objective function and appropriate restrictions on the variables:
Maximize
:
∑
t
=
1
T
(
∑
p
,
e
:
IE
e
,
t
=
1
k
∈
DisplayLad
(
e
)
Score
p
IDE
p
,
e
,
t
+
k
+
∑
p
,
e
:
IE
e
,
t
=
1
k
∈
AdLag
(
e
)
Score
p
IAE
p
,
e
,
t
+
k
)
where,
t: time indices;
E: set of events;
e: event in E;
p: index for products belonging to a set of promoted products;
DisplayLag
(
e
)
≡
{
the
offsets
from
the
start
of
the
event
during
which
the
products
have
to
be
on
display
}
;
AdLag
(
e
)
≡
{
the
offsets
from
the
start
of
the
event
during
which
the
products
have
to
be
advertised
}
;
T: total planning period in weeks;
IE
e
,
t
≡
{
1
if
event
e
is
scheduled
to
start
in
period
t
0
otherwise
;
IDE
p
,
e
,
r
≡
{
1
if
product
p
is
on
display
in
event
e
in
period
r
0
otherwise
;
IAE
p
,
e
,
r
≡
{
1
if
product
p
is
advertised
in
event
e
in
time
r
0
otherwise
;
and
r. any period in the promotion interval.
35. The computer-implemented method as recited in claim 28 , wherein the more than one manufacturer includes a plurality of manufacturers and wholesalers.