Shared data induced production process improvement

Patent metadata

Field	Value
Publication number	US-12596977-B2
Application number	US-202218282617-A
Country	US
Kind code	B2
Filing date	Mar 17, 2022
Priority date	Mar 19, 2021
Publication date	Apr 7, 2026
Grant date	Apr 7, 2026

Abstract

A method for developing or improving a process for producing a product from a material comprising steps of acquiring process data from at least two different sources for the production process and its relevant parameters by using a Data Collecting computer; using the acquired process data related to the production process to perform a Process Mapping step by using a Process Mapping computer; assigning the acquired process data related to the relevant parameters of the production process to its corresponding process parts by performing a Data Mapping step by using a Data Mapping computer; analyzing the therefore mapped process data with a specific software performed on an Analyzing computer thereby identifying and validating one or more existing characteristics related to the quality or performance of the production process; and using the identified and validated characteristics to develop the production process or improve its performance.

First claim

The invention claimed is:

1 . A method for improving a process for producing a product from a material comprising the following steps:
acquiring first process data from a first source of a plurality of sources for a production process of the product, second process data from a second source of the plurality of sources for the production process of the product, and parameters of the production process of the product,
wherein the first process data comprises first defectivity data of the production process at the first source, the second process data comprises second defectivity data of the production process at the second source, and the production process comprises a process part;
retrieving, based at least on the acquired first process data and the acquired second process data, a description of the production process from the first source and the second source; assigning the acquired first and second process data to respectively corresponding process part of the production process according to the parameters of the production process of the product; identifying and validating, based on the acquired first and second process data, by a trained machine learning model previously trained by using the description and defectivity data of a produced product, a characteristic of performance quality of the production process, wherein the characteristic indicates a removal rate of defective products from output of the production process; and creating, based on the identified and validated characteristic, at least the respectively corresponding process part of the production process, thereby improving the performance quality of the production process.

2 . The method of claim 1 , wherein the acquiring the first process data further comprises retrieving the first process data from a database, and the database has been created by observing the production process using data collecting devices comprising sensors and through an interactive user interface.

3 . The method of claim 2 , wherein the acquiring the first process data further comprises observing the production process during previous executions of the production process and/or during a current execution of the production process after using the identified and validated characteristic of the performance quality of the production process.

4 . The method of claim 1 , wherein the retrieving the description of the production process further comprises describing a structure of the production process or its pre-stages including necessary components, a process sequence, ingredients, and/or raw material.

5 . The method of claim 1 , wherein the assigning the acquired first and second process data further comprises assigning the acquired parameters, like temperature, mixing ratio of raw material, time, and the like, to corresponding process components and process sequences.

6 . The method of claim 1 , wherein the identifying and validating further uses supervised algorithms including a data analysis framework with a data model using approaches, the approaches comprises Multivariate Analysis, and the Multivariate Analysis comprises PLS regression, PCA, Random Forest, XGBoost, and an artificial neural network, using supervised and/or unsupervised algorithms.

7 . The method of claim 6 , wherein the XGBoost, the PLS model, the Random Forest, and the artificial neural network are for use as the supervised algorithms, and a structure of the supervised algorithms based on a result of training the artificial neural network according to the retrieved description of the production process.

8 . The method of claim 1 , wherein the acquiring the first process data further comprises examining the first and second sources either manually by a user or automatically performed by a remote computer through communication over a network.

9 . The method of claim 8 , wherein the plurality of sources comprises at least two distinct production sites.

10 . The method according to claim 9 , wherein the first process data comprises raw material data like specific quality parameters or metal impurity and purity levels, in-process-data comprising temperatures, pressures, flows, and/or P&ID charts.

11 . The method of claim 1 , wherein the characteristics comprises a root cause, like maintenance problems, or previously unknown process issues related to the performance quality of the production process comprising specific setting parameters for the production process.

12 . The method of claim 1 , wherein a new production process is created or an existing production process is improved by a user by setting up the new production process by using the identified characteristic or improving the existing production process by adapting improvement according to the identified characteristic.

13 . A System for developing or improving a process for producing a product from a material, the system comprises:
a processor; and a memory storing computer-executable instructions that when executed by the processor cause the system to execute operations comprising: acquiring first process data from a first production site of a plurality of production sites for a production process of the product, second process data from a second production site of the plurality of production sites for the production process of the product, and parameters of the production process of the product,
wherein the first process data comprises first defectivity data of the production process at the first production site, the second process data comprises second defectivity data of the production process at the second production site, and the production process comprises a process part;
retrieving, based at least on the acquired first process data and the acquired second process data, a description of the production process from the first production site and the second production site; assigning the acquired first and second process data to respectively corresponding process part of the production process according to the parameters of the production process of the product; identifying and validating, based on the acquired first and second process data, according to supervised algorithms including a data analysis framework with a data model, characteristic of performance quality of the production process,
wherein the characteristic indicates a removal rate of defective products from output of the production process; and
creating, based the identified and validated characteristic, at least the respectively corresponding process part of the production process, thereby improving the performance quality of the production process.

14 . The System according to claim 13 , wherein the first production site of the plurality of production sites represents a factory for producing at least one of chemicals or pharmaceuticals, and the second production site of the plurality of production sites represents a chemical material provider and/or distributor.

15 . The System according to claim 13 , wherein the acquiring the first process data is performed at least in part by a computer based digital platform.

16 . The System according to claim 13 , wherein
the retrieving the description of the production process is based at least in part on first interactive operations by a first operator, the assigning the acquired first and second process data is based at least in part on second interactive operations by a second operator, the identifying and validating the characteristic is performed by a server executing according to the supervised and/or unsupervised algorithms, notably an artificial neural network, and the crea

Assignees

• Versum Mat Us Llc

Inventors

• Kurt Safa Kutup
• Roth Rolf
• Matz Laura
• Wood Samuel

Classifications

• G06Q50/04

  Manufacturing · CPC title

• G06Q10/06395

  Quality analysis or management · CPC title

• G06Q10/06375

  Prediction of business process outcome or impact based on a proposed change · CPC title

• G06N3/08

  Learning methods · CPC title

• G06Q10/067

  Enterprise or organisation modelling · CPC title