# US 2024/0004935 A1 · Chemistry, Manufacturing, and Controls Change Assessment System

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0004935 A1
- Document ID: US20240004935A1
- Status: Published application
- Kind: A1
- Application number: 18/265,012
- Filed: 2021-11-30
- Published: 2024-01-04
- Earliest claimed: 2020-12-03
- Assignees: Merck Sharp & Dohme Corp.; Msd It Innovation Center S.R.O.
- Applicants: Merck Sharp & Dohme LLC; Msd It Innovation Center S.R.O.
- Inventors: John William Baker; Kyle Hayes; Jose Colon Colon; Regis Antonio Gimenis; Sachin Lalit Doshi; Jacob Rege Klitzke; Michael Bain; Pavlo Dvornikov; Nicolas Rey Turner; Peter Mezensky; Craig Kennedy; Mykyta Viazovskyi; Nikita Vostrosablin; Nirdosh Jagota; Marcy Newquist Huang
- Web page: https://patentgrep.com/patent/US20240004935A1

## CPC

- G06F 16/9038: Presentation of query results
- G06Q 50/04: Manufacturing
- G06F 16/254: {Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses}
- G06F 16/901: Indexing; Data structures therefor; Storage structures (for retrieval from the web G06F16/951)
- G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)}
- G06Q 10/06395: {Quality analysis or management}
- G06Q 10/0875: Itemisation or classification of parts, supplies or services, e.g. bill of materials
- G06Q 50/26: Government or public services (business processes related to the transportation industry G06Q50/40)
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A user (e.g., a CMC scientist) identifies a change in CMC data for a regulated product, service, or system provided by an organization. A change assessment engine provides a recommendation for regulatory impact reporting. The recommendation may include a rationale and supporting evidence. The recommendation may identify the markets where the product, service, or system is registered, an indication of potential reportability in that market, specific health authority regulations that apply, and direct linkages into an internal regulatory knowledge repository of the organization. The change assessment engine may drive a user interface that guides users through the process in a consistent and repeatable manner.

## Claims

### Claim 1

A computer-implemented change assessment method for a regulated product, system, or service, the method comprising:
receiving identification of a change proposal;
accessing change proposal data corresponding to the identified change proposal, the change proposal data having been ingested from a plurality of source datastores;
supplementing the change proposal data based on a user's responses to questions;
applying, to the change proposal data, a machine-learning model to generate one or more classifications;
providing, for display to the user, one or more matching criteria questions that are generated based on the classifications;
generating assessment results responsive to responses to the matching criteria questions; and
providing the assessment results for display to the user.

### Claim 2 (depends on 1)

The computer-implemented change assessment method of claim 1, further comprising ingesting the change proposal data from the plurality of source datastores.

### Claim 3 (depends on 2)

The computer-implemented change assessment method of claim 2, wherein ingesting the change proposal data comprises:
retrieving data relating to the change proposal from each of the plurality of source datastores;
converting the data relating to the change proposal into a standardized format to create a standardized change proposal record; and
storing the standardized change proposal record in a change assessment datastore that is distinct from the plurality of source datastores.

### Claim 4 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein the change proposal data includes at least one of: a description of the change, a reason for the change, technical information relating to the change, a comment on the change, an identification of a product impacted by the change, or an identification of a production site impacted by the change.

### Claim 5 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein the change proposal data is stored in a graph, the graph including nodes storing information about the regulated product and edges, between pairs of the nodes, indicating connections or correspondences between the information stored in corresponding pairs of nodes.

### Claim 6 (depends on 5)

The computer-implemented change assessment method of claim 5, wherein the change proposal is represented by a change proposal node in the graph, the change proposal node connected by one or more edges to at least one of: a family node indicating a family of the regulated product, a classification node indicating a classification of the change proposal, a material node indicating a material impacted by the change, or a material type node indicating a type of the material impacted by the change.

### Claim 7 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein the questions include at least one of: prompting the user to identify a material impacted by the change, prompting the user to assign a type to the material impacted by the change, or asking the user whether product quality will be impacted by the change.

### Claim 8 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein the machine-learning model is a natural language processing (NLP) model trained to generate recommended classifications for the change proposal based on text included in the change proposal data.

### Claim 9 (depends on 8)

The computer-implemented change assessment method of claim 8, wherein the NLP model is an Adaboost decision tree.

### Claim 10 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein providing the one or more matching criteria questions comprises:
identifying one or more regulations that are potentially relevant to the change;
providing, for display to the user, an indication of the one or more regulations; and
prompting the user to confirm whether the one or more regulations are relevant.

### Claim 11 (depends on 10)

The computer-implemented change assessment method of claim 10, wherein identifying the one or more regulations comprises querying a set of regulation mappings using the one or more classifications to identify the one or more regulations.

### Claim 12 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein providing the assessment results for display to the user comprises providing the user with access to text of one or more regulations determined to be relevant to the change.

### Claim 13 (depends on 1)

The computer-implemented change assessment method of claim 1, wherein the regulated product is a pharmaceutical product, a vaccine product, or a biological product.

### Claim 14

A non-transitory computer-readable storage medium comprising instructions for change assessment for a regulated product, service, or system, the instructions, when executed by a computing system, causing the computing system to:
receive identification of a change proposal;
access change proposal data corresponding to the identified change proposal, the change proposal data having been ingested from a plurality of source datastores;
supplement the change proposal data based on a user's responses to questions;
apply, to the change proposal data, a machine-learning model to generate one or more classifications;
provide, for display to the user, one or more matching criteria questions that are generated based on the classifications;
generate assessment results responsive to responses to the matching criteria questions; and
provide the assessment results for display to the user.

### Claim 15 (depends on 14)

The non-transitory computer-readable storage medium of claim 14, further comprising instructions that cause the computing system to:
retrieve data relating to the change proposal from each of the plurality of source datastores;
convert the data relating to the change proposal into a standardized format to create a standardized change proposal record; and
store the standardized change proposal record in a change assessment datastore that is distinct from the plurality of source datastores.

### Claim 16 (depends on 1)

The non-transitory computer-readable storage medium of claim 1, wherein the change proposal data is stored in a graph, the graph including nodes storing information about the regulated product and edges, between pairs of the nodes, indicating connections or correspondences between the information stored in corresponding pairs of nodes, and wherein the change proposal is represented by a change proposal node in the graph, the change proposal node connected by one or more edges to at least one of: a family node indicating a family of the regulated product, a classification node indicating a classification of the change proposal, a material node indicating a material impacted by the change, or a material type node indicating a type of the material impacted by the change.

### Claim 17 (depends on 1)

The non-transitory computer-readable storage medium of claim 1, wherein the instructions that cause the computing system to provide the one or more matching criteria questions comprise instructions that cause the computing system to:
identify one or more regulations that are potentially relevant to the change;
provide, for display to the user, an indication of the one or more regulations; and
prompt the user to confirm whether the one or more regulations are relevant.

### Claim 18 (depends on 17)

The non-transitory computer-readable storage medium of claim 17, wherein the instructions that cause the computing system to identify the one or more regulations comprise instructions that cause the computing system to query a set of regulation mappings using the one or more classifications to identify the one or more regulations.

### Claim 19 (depends on 14)

The non-transitory computer-readable storage medium of claim 14, wherein the instructions that cause the computing system to provide the assessment results for display to the user comprise instructions that cause the computing system to provide the user with access to text of one or more regulations determined to be relevant to the change.

### Claim 20

A change assessment system comprising:
one or more processors; and
a memory storing instructions that, when executed, cause the one or more processors to:
receive identification of a change proposal;
access change proposal data corresponding to the identified change proposal, the change proposal data having been ingested from a plurality of source datastores;
supplement the change proposal data based on a user's responses to questions;
apply, to the change proposal data, a machine-learning model to generate one or more classifications;
provide, for display to the user, one or more matching criteria questions that are generated based on the classifications;
generate assessment results responsive to responses to the matching criteria questions; and
provide the assessment results for display to the user.
