Sustainable resources exchange method and system

Patent metadata

Field	Value
Publication number	US-11830070-B2
Application number	US-202117411785-A
Country	US
Kind code	B2
Filing date	Aug 25, 2021
Priority date	Aug 25, 2021
Publication date	Nov 28, 2023
Grant date	Nov 28, 2023

Abstract

A system to facilitate the exchange and re-use of industrial waste and byproduct materials that have a reuse possibility is disclosed. The system promotes recycling, reusing, and reclaiming of wastes sustainably to ensure compliance and waste minimization. In one embodiment, the system supports an intelligent service for generating recommendations across an industrial stock exchange. Companies looking to dispose of their wastes or byproducts can access the system to initiate and execute trade with other companies looking for materials or compounds that could be reused in their own manufacturing processes. Both parties will be guided toward their shared goal in order to provide a net-zero impact on the environment and maintain an ecological balance.

First claim

We claim:

1. A computer-implemented method for applying machine learning to identify latent features of suppliers for an exchange of industrial by-products, the method comprising:
receiving, at a recommendation system, a dataset including, for multiple suppliers, their energy usage, carbon intensity, green practice certification, greenhouse gas emission, available manufacturing process capabilities, and available chemical compounds;
receiving, at the recommendation system, labeled data that includes data patterns representing correspondences between a desired chemical compound and manufacturing process capabilities that can be used to convert other chemical compounds to the desired chemical compound;
training a recommendation machine learning (ML) model for the recommendation system on the data patterns using a random-forest algorithm to determine a recommendation to present to a user in response to a set of feature values selected by the user, wherein the recommendation includes a transaction that has minimum energy emissions based on a distance between a supplier and the user;
receiving, from a first user and at the trained recommendation ML model, a set of feature values including a first chemical compound and manufacturing process capabilities available to the first user;
generating, via the trained recommendation ML model, a matrix identifying one or more potential suppliers, available chemical compounds that can be converted to the first chemical compound, and manufacturing process capabilities available to each of the potential suppliers, or to the first user, that can be used to convert each of the available chemical compounds to the first chemical compound;
calculating, for each of the one or more suppliers, a sustainability score based on the supplier's energy usage, carbon intensity, and greenhouse gas emission;
determining, via a decision tree algorithm of the trained recommendation ML model, whether the sustainability score for each supplier is greater than a selected classification threshold;
filtering the matrix to exclude suppliers whose sustainability score is below the classification threshold; and
presenting on a computing device, via a user interface of the recommendation system and based only on the suppliers included in the filtered matrix, a recommendation to the first user including either or both of:
direct suppliers offering the first chemical compound, or
indirect suppliers offering another chemical compound that can be converted to the first chemical compound via manufacturing process capabilities available to the indirect supplier and/or the first user.

2. The method of claim 1 , further comprising:
receiving, from the first user and at the recommendation system, one or more product fulfillment parameters as part of the set of feature values; and
further filtering, via the recommendation system, the matrix to include only suppliers that match the first user's product fulfillment parameters.

3. The method of claim 2 , wherein the product fulfillment parameters include one or more of industry type, supplier company rating, location of supplier, distance from the first user to supplier, manufacturing processes available to the supplier, cost per tonne, minimum order quantity, and types of waste.

4. The method of claim 1 , wherein the sustainability score for a supplier is calculated using a first calculation based on the supplier's green practice certification and energy usage, and a second calculation based on the supplier's carbon intensity and greenhouse gas emission.

5. The method of claim 1 , wherein the one or more suppliers include a first supplier, and the method further comprises:
calculating, at the recommendation system, a first sustainability score for the first supplier;
determining, at the recommendation system, that the first sustainability score is below the selected classification threshold; and
excluding, at the recommendation system, the first supplier from the recommendation.

6. The method of claim 5 , further comprising displaying, at a supplier dashboard of the recommendation system for the first supplier, an alert showing the first sustainability score compared to a target sustainability score.

7. The method of claim 5 , further comprising displaying, at a supplier dashboard of the recommendation system for the first supplier, an alert describing the exclusion, and one or more recommendations for improving their sustainability score.

8. The method of claim 1 , further comprising presenting in the recommendation, one or more industry-specific trends related to re-usability of by-products in order to promote a circular economy.

9. The method of claim 1 , wherein the recommendation includes information about each supplier's sustainability.

10. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to apply machine learning to identify latent features of suppliers for an exchange of industrial by-products by performing the following:
receive, at a recommendation system, a dataset including, for multiple suppliers, their energy usage, carbon intensity, green practice certification, greenhouse gas emission, available manufacturing process capabilities, and available chemical compounds;
receive, at the recommendation system, labeled data that includes data patterns representing correspondences between a desired chemical compound and manufacturing process capabilities that can be used to convert other chemical compounds to the desired chemical compound;
train a recommendation machine learning (ML) model for the recommendation system on the data patterns using a random-forest algorithm to determine a recommendation to present to a user in response to a set of feature values selected by the user, wherein the recommendation includes a transaction that has minimum energy emissions based on a distance between a supplier and the user;
receive, from a first user and at the trained recommendation ML model, a set of feature values including a first chemical compound and manufacturing process capabilities available to the first user;
generate, via the trained recommendation ML model, a matrix identifying one or more potential suppliers, available chemical compounds that can be converted to the first chemical compound, and manufacturing process capabilities available to each of the potential suppliers, or to the first user, that can be used to convert each of the available chemical compounds to the first chemical compound;
calculate, for each of the one or more suppliers, a sustainability score based on the supplier's energy usage, carbon intensity, and greenhouse gas emission;
determine, via a decision tree algorithm of the trained recommendation ML model, whether the sustainability score for each supplier is greater than a selected classification threshold;
filter the matrix to exclude suppliers whose sustainability score is below the classification threshold; and
present on a computing device, via a user interface of the recommendation system and based only on the suppliers included in the filtered matrix, a recommendation to the first user including either or both of:
direct suppliers offering the first chemical compound, or
indirect suppliers offering another chemical compound that can be converted to the first chemical compound via manufacturing process capabilities available to the indirect supplier and/or the first user.

11. The non-transitory computer-readable medium storing software of claim 10 , wherein the instructions further cause the one or more computers to:
receive, from the first user and at the recom

Assignees

• Accenture Global Solutions Ltd

Inventors

• Raj Sahil
• Kumar Mithilesh
• Tiwari Saumya

Classifications

• G06Q40/04 Primary

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title

• G06Q30/0201 Primary

  Market modelling; Market analysis; Collecting market data · CPC title

• G06Q30/0283

  Price estimation or determination · CPC title

• G06F18/24

  Classification techniques · CPC title