System for selecting an optimal sample set of jobs for determining price models for a print market port
View Patent ↗A system for determining price models of a print market port including a processor and a computer-readable storage medium in communication with the processor, wherein the computer-readable storage medium comprises one or more programming instructions for: partitioning a job dataset into a plurality of categories, each of the plurality of categories having a pricing model; determining one or more factors within the job dataset that influence a price of each job; developing an input/output model for each job in the job dataset that influences the price of the job; performing an iteration to compute a prediction error for each job in the job dataset that influences the price of the job; removing one or more jobs from a subsequent iteration that include prediction errors that exceed a prediction error threshold; and performing a plurality of iterations on remaining jobs until a predetermined average error prediction is reached.
1. A system for determining price models of a print market port, the system comprising:
processor; and
a computer-readable storage medium in communication with said processor, wherein the computer-readable storage medium comprises one or more programming instructions for:
partitioning a job dataset into a plurality of categories, each of the plurality of categories having a pricing model;
determining one or more factors within the job dataset that influence a price of each job in the job dataset;
developing an input/output model for each job in the job dataset that influences the price of the job;
performing an iteration to compute a prediction error for each job in the job dataset that influences the price of the job;
removing one or more jobs from a subsequent iteration that include prediction errors that exceed a prediction error threshold;
performing a plurality of iterations on remaining jobs until a predetermined average error prediction is reached;
determining an optimal job dataset including a sample set of jobs remaining after the plurality of iterations; and, determining the pricing model using the optimal job dataset.
2. The system according to claim 1 , wherein the job dataset includes historical job data.
3. The system according to claim 1 , wherein the input/output model is developed by utilizing a neural network model.
4. The system according to claim 1 , wherein the price of the job is determined as a function of job attributes that influence the price of the job.
5. The system according to claim 1 , wherein the plurality of categories include pre-press, pre-press and print, book, envelope, open item, commercial print, and direct mail.
6. The system according to claim 1 , wherein the prediction error is computed by comparing an actual price of each job to a predicted price of each job to determine an absolute error and an absolute percentage error for each job.
7. The system according to claim 1 , wherein the predetermined average error prediction is a constant.
8. The system according to claim 7 , wherein the constant is 2%.
9. The system according to claim 1 , wherein the predetermined average error prediction is based on a median number derived from a Gaussian distribution calculation.
10. The system according to claim 1 , wherein select jobs of the job dataset are manually entered into the system.
11. A system for selecting an optimal set of jobs, the system comprising:
a processor; and
a computer-readable storage medium in communication with said processor, wherein the computer-readable storage medium comprises one or more programming instructions for:
collecting historical job data related to a plurality of jobs;
analyzing the historical job data based upon one or more criteria;
constructing a neural network model for each of the plurality of jobs;
predicting a cost of each of the plurality of jobs;
computing a cost prediction error by comparing the cost predicted for each of the plurality of jobs to an actual cost for each of the plurality of jobs; and
iteratively removing one or more jobs of the plurality of jobs having the cost prediction error greater than a predetermined cost prediction error until a predetermined average error prediction is reached; and
determining a pricing model using an optimal set of jobs remaining after the predetermined average error prediction is reached.
12. The system according to claim 11 , wherein the cost of each of the plurality of jobs is determined as a function of job attributes that influence the price of each job of the plurality of jobs.
13. The system according to claim 11 , wherein the historical job data is partitioned into plurality of categories that each include pre-press, pre-press and print, book, envelope, open item, commercial print, and direct mail.
14. The system according to claim 11 , wherein the comparing step determines an absolute error and an absolute percentage error for each job of the plurality of jobs.
15. The system according to claim 11 , wherein the predetermined cost prediction error is a constant.
16. The system according to claim 15 , wherein the constant is 2%.
17. The system according to claim 11 , wherein the predetermined cost prediction error is based on a median number derived from a Gaussian distribution calculation.
18. The system according to claim 11 , wherein select jobs of the plurality of jobs are manually entered into the system.
19. The system according to claim 11 , wherein the system further determines price models for a print market port.
20. The system according to claim 11 , wherein the system is automated.