System and method for disaggregating weekly forecast into daily components
A computer-implemented method transforms transactional data into a forecast of demand for controlling a commerce system. Goods move between members of a commerce system. The transactional data is recorded. Daily disaggregating parameters (DDP) are estimated by minimizing an error function of day of week and transactional data grouped according to promotion, price range, or customer. Model parameters are estimated based on the DDP and transactional data using a demand model to generate a weekly forecast of demand for the good. The weekly forecast of demand is disaggregated into daily components using the DDP. The daily components of the forecast of demand are provided to a member of the commerce system to control the movement of goods in the commerce system. Any variation in the demand model due to promotions, price changes, out-of-stock, and low selling product is taken into account when determining the daily components.
1. A computer-implemented method for transforming transactional data into a forecast of demand for controlling a commerce system, comprising:
recording transactional data related to a movement of goods between members of a commerce system, the transactional data including price, product, time, promotion, and customer;
estimating, by a processor, daily disaggregating parameters (DDP) by minimizing an error function of day of week and transactional data grouped according to promotion, price range, or customer;
estimating, by a processor, model parameters based on the DDP and transactional data using a demand model to generate a weekly forecast of demand for the goods;
disaggregating, by a processor, the weekly forecast of demand into daily components using the DDP;
determining, by a processor, the daily components of the weekly forecast of demand by minimizing the error function given by:
error
(
k
,
d
,
t
)
=
Txn
(
k
,
t
)
(
w
(
d
)
-
w
act
(
k
,
d
,
t
)
(
∑
d
ε
/
(
k
,
t
)
w
(
d
)
)
)
;
and
providing the daily components of the weekly forecast of demand for the goods to a member of the commerce system to control the movement of goods in the commerce system.
2. The computer-implemented method of claim 1 , wherein the error function is subject to constraints given by:
∑
d
=
1
7
w
(
d
)
=
1
,
and
w
(
d
)
≥
0
,
for
d
=
1
,
…
,
7.
3. The computer-implemented method of claim 1 , further including accounting, by a processor, for variation in the demand model due to promotions, price changes, and out-of-stock of a product when determining the daily components of the weekly forecast of demand.
4. The computer-implemented method of claim 1 , further including accounting, by a processor, for variation in the demand model due to low selling products when determining the daily components of the weekly forecast of demand.
5. The computer-implemented method of claim 1 , wherein disaggregating the weekly forecast for demand into daily components of the weekly forecast of demand uses less computing resources than daily aggregation.
6. A computer-implemented method for controlling commercial transactions using a forecast of demand, comprising:
recording transactional data related to a movement of goods between members of a commerce system;
estimating, by a processor, daily disaggregating parameters (DDP) by minimizing an error function of day of week and transactional data grouped according to promotion, price range, or customer;
estimating, by a processor, model parameters based on the DDP and transactional data using a demand model to generate a forecast of demand for the goods based on a first time scale;
disaggregating, by a processor, the forecast of demand into components based on a second time scale using the DDP;
determining, by a processor, the components of the forecast of demand by minimizing the error function given by:
error
(
k
,
d
,
t
)
=
Txn
(
k
,
t
)
(
w
(
d
)
-
w
act
(
k
,
d
,
t
)
(
∑
d
∈
I
(
k
,
t
)
w
(
d
)
)
)
;
providing the components of the forecast of demand for the goods to a member of the commerce system to control the movement of goods in the commerce system.
7. The computer-implemented method of claim 6 , wherein the first time scale is weekly and the second time scale is daily.
8. The computer-implemented method of claim 6 , wherein the error function is subject to constraints given by:
∑
d
=
1
7
w
(
d
)
=
1
,
and
w
(
d
)
≥
0
,
for
d
=
1
,
…
,
7.
9. The computer-implemented method of claim 6 , further including accounting, by a processor, for variation in the demand model due to promotions, price changes, and out-of-stock of a product when determining the components of the forecast of demand.
10. The computer-implemented method of claim 6 , further including accounting, by a processor, for variation in the demand model due to low selling products when determining the components of the forecast of demand.
11. The computer-implemented method of claim 6 , wherein disaggregating the forecast for demand into daily components uses less computing resources than daily aggregation.
12. A computer-readable medium having computer readable program code embodied thereon and capable of execution by a computer, the medium comprising:
program code to record transactional data related to a movement of goods between members of a commerce system;
program code to estimate daily disaggregating parameters (DDP) by minimizing an error function of day of week and transactional data grouped according to promotion, price range, or customer;
program code to estimate model parameters based on the DDP and transactional data using a demand model to generate a forecast of demand for the goods based on a first time scale;
program code to disaggregate the forecast of demand into components based on a second time scale using the DDP
program code to determine the components of the forecast of demand by minimizing the error function given by:
error
(
k
,
d
,
t
)
=
Txn
(
k
,
t
)
(
w
(
d
)
-
w
act
(
k
,
d
,
t
)
(
∑
d
∈
I
(
k
,
t
)
w
(
d
)
)
)
;
and
program code to provide the components of the forecast of demand for the goods to a member of the commerce system to control the movement of goods in the commerce system.
13. The medium of claim 12 , wherein the first time scale is weekly and the second time scale is daily.
14. The medium of claim 12 , further including program code to account for variation in the demand model due to promotions, price changes, and out-of-stock of a product when determining the components of the forecast of demand.
15. The medium of claim 12 , further including program code to account for variation in the demand model due to low selling products when determining the components of the forecast of demand.
16. The medium of claim 12 , wherein disaggregating the forecast for demand into daily components uses less computing resources than daily aggregation.
17. A system for controlling commercial transactions using a forecast of demand, comprising:
a commerce system having a plurality of members each having a control system for controlling movement of goods between the members;
a database in electronic communication with the commerce system for recording transactional data related to movement of goods between the members of the commerce system; and
a computer in electronic communication with the commerce system for estimating daily disaggregating parameters (DDP) by minimizing an error function of day of week and transactional data grouped according to promotion, price range, or customer, estimating model parameters based on the DDP and transactional data using a demand model to generate a forecast of demand for the goods based on a first time scale, and disaggregating the forecast of demand into components based on a second time scale using the DDP;
wherein the components of the forecast of demand is determined by minimizing the error function given by:
error
(
k
,
d
,
t
)
=
Txn
(
k
,
t
)
(
w
(
d
)
-
w
act
(
k
,
d
,
t
)
(
∑
d
∈
I
(
k
,
t
)
w
(
d
)
)
)
;
and
wherein the components of the forecast of demand for the goods is provided to a member of the commerce system to control the movement of goods in the commerce system.
18. The system of claim 17 , wherein the first time scale is weekly and the second time scale is daily.
19. The system of claim 17 , wherein variation in the demand model due to promotions, price changes, and out-of-stock of a product is taken into account when determining the components of the forecast of demand.
20. The system of claim 17 , wherein variation in the demand model due to low selling products is taken into account when determining the components of the forecast of demand.
21. The system of claim 17 , wherein disaggregating the forecast for demand into daily components uses less computing resources than daily aggregation.