# US 2024/0005128 A1 · Prediction method, information processing apparatus, film forming apparatus, article manufacturing method and non-transitory storage medium

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

- Number: US 2024/0005128 A1
- Document ID: US20240005128A1
- Status: Published application
- Kind: A1
- Application number: 18/336,149
- Filed: 2023-06-16
- Published: 2024-01-04
- Earliest claimed: 2022-07-01
- Assignees: Canon Kabushiki Kaisha
- Applicants: Canon Kabushiki Kaisha
- Inventors: NAOKI KIYOHARA; SENTARO AIHARA; YUICHIRO OGUCHI
- Web page: https://patentgrep.com/patent/US20240005128A1

## CPC

- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/042: Knowledge-based neural networks; Logical representations of neural networks
- G06N 3/08: Learning methods
- G06T 7/0004: {Industrial image inspection}
- G06N 3/048: Activation functions
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A prediction method of predicting a behavior of droplets of a curable composition in a process of forming a film of the curable composition from a plurality of droplets of the curable composition arranged on a first member, the method including predicting the behavior of the droplets using a learning model, wherein an input of the learning model includes first information indicating positions on the first member to which the droplets of the curable composition are to be arranged.

## Claims

### Claim 1

A prediction method of predicting a behavior of droplets of a curable composition in a process of forming a film of the curable composition from a plurality of droplets of the curable composition arranged on a first member, the method comprising
predicting the behavior of the droplets using a learning model,
wherein an input of the learning model includes first information indicating positions on the first member to which the droplets of the curable composition are to be arranged.

### Claim 2 (depends on 1)

The method according to claim 1, wherein
the input of the learning model includes second information indicating a relative positional relationship among adjacent droplets of the plurality of droplets.

### Claim 3 (depends on 2)

The method according to claim 2, wherein
the second information is expressed by a graph formed from nodes and a link connecting the nodes.

### Claim 4 (depends on 3)

The method according to claim 3, wherein
in the graph, a position corresponding to each of the plurality of droplets is set as the node, and a line connecting adjacent nodes is set as the link.

### Claim 5 (depends on 4)

The method according to claim 4, wherein
the position includes a center position of each of the plurality of droplets.

### Claim 6 (depends on 3)

The method according to claim 3, wherein
in the graph, a position in a region defined by adjacent droplets of the plurality of droplets is set as the node, and a line connecting adjacent nodes is set as the link.

### Claim 7 (depends on 6)

The method according to claim 6, wherein
the region is a region divided by Voronoi boundaries of a Voronoi diagram in which each of the plurality of droplets is set as a generating point.

### Claim 8 (depends on 6)

The method according to claim 6, wherein
the position includes a centroid position of the region.

### Claim 9 (depends on 6)

The method according to claim 6, wherein
the position includes a Voronoi boundary of a Voronoi diagram in which each of the plurality of droplets is set as a generating point.

### Claim 10 (depends on 3)

The method according to claim 3, wherein
in the graph, a position corresponding to each of the plurality of droplets, a position in a region defined by adjacent droplets of the plurality of droplets, and a position where Voronoi boundaries of a Voronoi diagram intersect, in which each of the plurality of droplets is set as a generating point, are set as nodes, and a line connecting adjacent nodes is set as the link.

### Claim 11 (depends on 3)

The method according to claim 3, wherein
in the graph, at least one of the node and the link holds, as a feature amount, one of information concerning a position and a volume of the droplet and information concerning a shape of the first member.

### Claim 12 (depends on 2)

The method according to claim 2, further comprising
generating the learning model while using at least one of the first information and the second information as the input and using the predicted behavior of the droplets as learning data.

### Claim 13 (depends on 1)

The method according to claim 1, wherein
in the predicting, a degree of merging of adjacent droplets of the plurality of droplets is predicted as the behavior of the droplets.

### Claim 14 (depends on 1)

The method according to claim 1, wherein
the process includes a process of bringing the curable composition arranged on the first member and a second member into contact with each other, thereby forming a film of the curable composition in a space between the first member and the second member.

### Claim 15 (depends on 14)

The method according to claim 14, wherein
the input of the learning model includes at least one of information indicating a volume of the droplet, information concerning volatilization of the droplet, information concerning a shape of the first member, and information concerning a shape of a pattern provided on the second member.

### Claim 16 (depends on 2)

The method according to claim 2, wherein
the learning model uses a graph neural network.

### Claim 17 (depends on 1)

The method according to claim 1, wherein
the learning model uses a neural network.

### Claim 18

An information processing apparatus that predicts a behavior of droplets of a curable composition in a process of forming a film of the curable composition from a plurality of droplets of the curable composition arranged on a first member, wherein
the apparatus predicts the behavior of the droplets using a learning model, and
an input of the learning model includes information indicating positions on the first member to which the droplets of the curable composition are to be arranged.

### Claim 19 (depends on 18)

A film forming apparatus incorporating an information processing apparatus defined in claim 18, wherein
a process of forming a film of a curable composition from a plurality of droplets of the curable composition arranged on a first member is controlled based on prediction of a behavior of the droplets of the curable composition by the information processing apparatus.

### Claim 20 (depends on 1)

An article manufacturing method comprising:
determining, while repeating a prediction method defined in claim 1, a condition for a process of forming a film of a curable composition from a plurality of droplets of the curable composition arranged on a first member, and
executing the process in accordance with the condition.

### Claim 21 (depends on 1)

A non-transitory storage medium storing a program for causing a computer to execute a prediction method defined in claim 1.
