Proactively predicting large orders and providing fulfillment support related thereto
Methods, apparatus, and processor-readable storage media for proactively predicting large orders and providing fulfillment support related thereto are provided herein. An example computer-implemented method includes classifying, via a first set of one or more machine learning techniques, a transaction quote as a transaction quote that exceeds one or more size-related parameters or a transaction quote that does not exceed the one or more size-related parameters; determining, if the transaction quote is classified as a transaction quote that exceeds one or more size-related parameters, supportability of converting the transaction quote into a transaction order via a second set of one or more machine learning techniques; and outputting, based on the determined supportability, information pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order to one or more entities associated with transaction order fulfillment.
1. A computer-implemented method comprising:
classifying a transaction quote as a transaction quote that exceeds one or more size- related parameters or a transaction quote that does not exceed the one or more size-related parameters, wherein classifying comprises determining dependencies based at least in part on one or more trends and seasonality information related to the transaction quote by processing the transaction quote and variables related to the transaction quote using a first set of one or more machine learning techniques comprising a recurrent neural network which includes one or more long short-term memory models;
training, using (i) a log function based at least in part on an exponential function and an inverse probability and (ii) historical attribute data pertaining to at least stock keeping unit information, geographic information, temporal information, customer information, quote quantity information, and sales representative information, at least one trained multivariate logistic regression algorithm for determining coefficients of attribute contribution to supportability of converting a transaction quote into a transaction order;
determining, if the transaction quote is classified as a transaction quote that exceeds one or more size-related parameters, supportability of converting the transaction quote into a transaction order, wherein determining supportability comprises applying a second set of one or more machine learning techniques to multiple attributes related to converting the transaction quote into a transaction order, wherein applying the second set of one or more machine learning techniques comprises determining levels of contributions to supportability for the multiple attributes by processing data related to the multiple attributes using at least one support vector machine model to calculate distances from a mean for the multiple attributes and (ii) the at least one trained multivariate logistic regression algorithm to determine a coefficient of each attribute contribution, across the multiple attributes, to supportability of converting the transaction quote into a transaction order; and
outputting, based on the determined supportability, information pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order to one or more entities associated with transaction order fulfillment;
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.
2. The computer-implemented method of claim 1 , wherein classifying the transaction quote comprises using a categorical boosting algorithm.
3. The computer-implemented method of claim 1 , wherein the multiple attributes comprise historical transaction order supportability data.
4. The computer-implemented method of claim 1 , wherein the multiple attributes comprise inventory availability.
5. The computer-implemented method of claim 1 , wherein the one or more entities associated with order fulfillment comprise one or more sales representatives.
6. The computer-implemented method of claim 1 , further comprising:
generating a confidence score pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order, and wherein the information comprises the generated confidence score.
7. The computer-implemented method of claim 1 , wherein the one or more entities associated with order fulfillment comprise one or more planners, and wherein the information comprises information regarding procurement of one or more components of the transaction order required for fulfilling the transaction order.
8. The computer-implemented method of claim 1 , wherein the information comprises information pertaining to adjusting a demand threshold.
9. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes said at least one processing device:
to classify a transaction quote as a transaction quote that exceeds one or more size-related parameters or a transaction quote that does not exceed the one or more size-related parameters, wherein classifying comprises determining dependencies based at least in part on one or more trends and seasonality information related to the transaction quote by processing the transaction quote and variables related to the transaction quote using a first set of one or more machine learning techniques comprising a recurrent neural network which includes one or more long short-term memory models;
to train, using (i) a log function based at least in part on an exponential function and an inverse probability and (ii) historical attribute data pertaining to at least stock keeping unit information, geographic information, temporal information, customer information, quote quantity information, and sales representative information, at least one trained multivariate logistic regression algorithm for determining coefficients of attribute contribution to supportability of converting a transaction quote into a transaction order;
to determine, if the transaction quote is classified as a transaction quote that exceeds one or more size-related parameters, supportability of converting the transaction quote into a transaction order, wherein determining supportability comprises applying a second set of one or more machine learning techniques to multiple attributes related to converting the transaction quote into a transaction order, wherein applying the second set of one or more machine learning techniques comprises determining levels of contributions to supportability for the multiple attributes by processing data related to the multiple attributes using at least one support vector machine model to calculate distances from a mean for the multiple attributes and (ii) the at least one trained multivariate logistic regression algorithm to determine a coefficient of each attribute contribution, across the multiple attributes, to supportability of converting the transaction quote into a transaction order; and
to output, based on the determined supportability, information pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order to one or more entities associated with transaction order fulfillment.
10. The non-transitory processor-readable storage medium of claim 9 , wherein classifying the transaction quote comprises using a categorical boosting algorithm.
11. An apparatus comprising:
at least one processing device comprising a processor coupled to a memory;
said at least one processing device being configured:
to classify a transaction quote as a transaction quote that exceeds one or more size-related parameters or a transaction quote that does not exceed the one or more size-related parameters, wherein classifying comprises determining dependencies based at least in part on one or more trends and seasonality information related to the transaction quote by processing the transaction quote and variables related to the transaction quote using a first set of one or more machine learning techniques comprising a recurrent neural network which includes one or more long short-term memory models;
to train, using (i) a log function based at least in part on an exponential function and an inverse probability and (ii) historical attribute data pertaining to at least stock keeping unit information, geographic information, temporal information, customer information, quote quantity information, and sales representative information, at least one trained multivariate logistic regression algorithm for determining coefficients of attribute contribution to supportability of converting a transaction quote into a transaction order;
to determine, if the transaction quote is classified as a transaction quote that exceeds one or more size-related parameters, supportability of converting the transaction quote into a transaction order, wherein determining supportability comprises applying a second set of one or more machine learning techniques to multiple attributes related to converting the transaction quote into a transaction order, wherein applying the second set of one or more machine learning techniques comprises determining levels of contributions to supportability for the multiple attributes by processing data related to the multiple attributes using (i) at least one support vector machine model to calculate distances from a mean for the multiple attributes and (ii) the at least one trained multivariate logistic regression algorithm to determine a coefficient of each attribute contribution, across the multiple attributes, to supportability of converting the transaction quote into a transaction order; and
to output, based on the determined supportability, information pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order to one or more entities associated with transaction order fulfillment.
12. The apparatus of claim 11 , wherein classifying the transaction quote comprises using a categorical boosting algorithm.
13. The apparatus of claim 11 , wherein the one or more entities associated with order fulfillment comprise one or more planners, and wherein the information comprises information regarding procurement of one or more components of the transaction order required for fulfilling the transaction order.
14. The non-transitory processor-readable storage medium of claim 9 , wherein the one or more entities associated with order fulfillment comprise one or more planners, and wherein the information comprises information regarding procurement of one or more components of the transaction order required for fulfilling the transaction order.
15. The non-transitory processor-readable storage medium of claim 9 , wherein the program code when executed by the at least one processing device causes the at least one processing device:
to generate a confidence score pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order, and wherein the information comprises the generated confidence score.
16. The apparatus of claim 11 , wherein the at least one processing device is further configured:
to generate a confidence score pertaining to converting the transaction quote into a transaction order and fulfilling the transaction order, and wherein the information comprises the generated confidence score.
17. The apparatus of claim 11 , wherein the multiple attributes comprise historical transaction order supportability data.
18. The apparatus of claim 11 , wherein the multiple attributes comprise inventory availability.
19. The non-transitory processor-readable storage medium of claim 9 , wherein the multiple attributes comprise historical transaction order supportability data.
20. The non-transitory processor-readable storage medium of claim 9 , wherein the multiple attributes comprise inventory availability.