# US 2026/0244825 A1 · Reliability assessment of an electronic component carrier or a package comprising it using ai-supported finite elements analysis

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

- Number: US 2026/0244825 A1
- Document ID: US20260244825A1
- Status: Published application
- Kind: A1
- Application number: 19/121,173
- Filed: 2023-10-16
- Published: 2026-08-20
- Earliest claimed: 2022-10-17
- Assignees: AT&S Austria Technologie & Systemtechnik Aktiengesellschaft
- Applicants: AT & S Austria Technologie & Systemtechnik Aktiengesellschaft
- Inventors: Thomas KRIVEC; Qi TAO; Julia ZUENDEL
- Web page: https://patentgrep.com/patent/US20260244825A1

## CPC

- G06F 30/23: using finite element methods [FEM] or finite difference methods [FDM]
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06F 30/17: Mechanical parametric or variational design
- G06F 2113/18: Chip packaging
- G06F 2115/12: Printed circuit boards [PCB] or multi-chip modules [MCM]
- G06F 2119/02: Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A reliability assessment method for electronic component carriers, the method comprising: providing a global simulation model of the component carrier, wherein geometric properties of the component carrier and its constituent parts are defined as independent variables in a parameter space; and the global simulation model receives input data including a point in the parameter space and outputs data indicative of resulting geometric and/or physical properties of the resulting component carrier. The method comprises: identifying boundary conditions for local sub-portions of the component carrier; providing data-based local models for said local sub-portions, each data-based local model being trained to receive boundary conditions as input data and provide a criticality value indicative of a reliability of the respective local sub-portion as output data; running the local data-based model for each local sub-portions; and identifying, in the respective output data, those local sub-portions whose criticality value is above at least one predetermined threshold.

## Claims

### Claim 1

A method for a reliability assessment of an electronic component carrier or a package comprising an electronic component carrier, the method comprising:
providing a global simulation model of the component carrier or package wherein a predetermined number of geometric properties of the component carrier or package and of its constituent parts made of different materials and of physical properties of materials used in said constituent parts are defined as independent variables in a parameter space of the global simulation model;
wherein the global simulation model is configured to receive input data including a point in the parameter space and is configured to output data indicative of resulting geometric and/or physical properties within a global volume of the resulting component carrier or package;
running a simulation of the global simulation model;
identifying, in the output data of the global simulation model, boundary conditions for a plurality of local sub-portions of the component carrier or package
providing at least one data-based local model for said local sub-portions, each data-based local model being trained so as to receive said boundary conditions as input data and provide a criticality value indicative of a reliability of the respective local sub-portion as output data; and
running the at least one data-based local model for each of said local sub-portions and identifying, in the respective output data, those local sub-portions whose criticality value is above at least one predetermined threshold.

### Claim 2 (depends on 1)

The method of claim 1, wherein:
said plurality of local sub-portions of the component carrier or package includes local sub-portions which belong to one or more of the following or other potentially critical types:
single laser vias;
stacks of at least two laser vias;
plated through holes;
non-plated through holes;
pads;
conductive lines or segments thereof;
solder bumps or solder balls;
copper pillars;
electronic components; and
wherein local sub-portions of these and/or other different types are preferably described by data-based local models of correspondingly different types.

### Claim 3 (depends on 1)

The method of claim 1, wherein said electronic component carrier or package comprising an electronic component carrier is one or more of the following:
a single-layer or multi-layer printed circuit board;
a high-density interconnect printed circuit board;
an assembled printed circuit board including at least one active or passive component;
an electronic module including mechanical components;
an IC substrate;
a single-or multi-die package;
a heterogeneous package; or
an electronic system combining one or more of the above-mentioned entities in one functional unit.

### Claim 4 (depends on 1)

The method of claim 1, wherein:
the global simulation model is based on a finite element method; and
the boundary condition for at least one local sub-portion comprises a boundary portion of a mesh used in the finite element method.

### Claim 5 (depends on 1)

The method of claim 1, wherein:
the at least one data-based local model is trained by a machine learning method based on at least one of:
an artificial (deep) neuronal net;
a Bayesian Algorithm; or
a Clustering Algorithm.

### Claim 6 (depends on 1)

The method of claim 1, wherein:
providing the at least one data-based local model and/or running the at least one data-based local model comprises providing and running a supplemented global simulation model which additionally includes a local simulation model for the respective local sub-portion.

### Claim 7 (depends on 1)

The method of claim 1, wherein:
the global simulation model is time-dependent in that said simulation run comprises a dynamic process including chemical and/or physical changes of the component carrier or package in particular under simulated varying environmental conditions whose initial values and/or variation rate are comprised in the input data of the global simulation model.

### Claim 8 (depends on 1)

The method of claim 1, wherein:
the output data of the global simulation model is time-dependent in that it comprises at least one function of time characterizing a dynamic processing of the component carrier or package including chemical and/or physical changes of the component carrier, in particular under varying environmental conditions whose initial values and/or variation rate are comprised in the input data of the global simulation model.

### Claim 9 (depends on 1)

The method of claim 1, wherein:
at least one of said data-based local models is time-dependent in that it comprises a dynamic process including chemical and/or physical changes of the respective local sub-portion of the component carrier or package, in particular under varying environmental conditions whose initial values and/or variation rate are comprised in the input data of the respective data-based local model.

### Claim 10 (depends on 1)

The method of claim 1, wherein:
for at least one of said data-based local models, additional geometric, physical and/or process parameters of the respective local sub-portion of the component carrier or package are comprised in the input data in addition to said boundary conditions obtained from the output of the global simulation model.

### Claim 11 (depends on 1)

The method of claim 1, wherein:
for at least one of said data-based local models, training input data comprise predefined geometric and/or physical deviations from a predetermined standard configuration of the respective local sub-portion of the component carrier or package; and
the criticality value in the corresponding training output data for said data-based local model is defined as a function of said deviation.

### Claim 12 (depends on 1)

The method of claim 1, further comprising:
calculating a global criticality value of the component carrier or package as a function of all the criticality values obtained from the local sub-portions; and
iteratively optimizing said global criticality value of the component carrier or package by selecting at least one of said local sub-portions and varying its geometric, physical and/or process parameters included in the input data of its data-based local model so as to reduce the calculated global criticality value of the component carrier or package until a minimum or another predefined optimal value of the global criticality value is achieved.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
manufacturing said component carrier or package depending on whether local sub-portions have been identified whose criticality values are above the at least one predetermined threshold;
adapting geometric, physical and/or process parameters of those local sub-portions of the component carrier or package, whose criticality values are identified to lie above the at least one predetermined threshold, such that their criticality values are reduced below the at least one predetermined threshold, and subsequently manufacturing the component carrier or package; or
manufacturing said component carrier or package according to those parameters of said local sub-portions of the component carrier or package, for which said minimum or said predefined optimal value of the global criticality value is achieved.

### Claim 14 (depends on 1)

A computer-readable medium comprising a memory storing a computer program for a reliability assessment of an electronic component carrier or package comprising an electronic component carrier, which computer program, when being executed by a processor, is adapted to carry out the method of claim 1.

### Claim 15

(canceled)

### Claim 16 (depends on 1)

A reliability assessment device adapted for performing reliability assessment of an electronic component carrier or package comprising an electronic component carrier, the reliability assessment device comprising a memory storing a computer program, which when being executed by a processor of the reliability assessment device is adapted to carry out the method of claim 1.

### Claim 17 (depends on 7)

The method of claim 7, wherein the dynamic process includes a heat-up and/or cool-down process.

### Claim 18 (depends on 8)

The method of claim 8, wherein the dynamic processing includes a heat-up and/or cool-down process.

### Claim 19 (depends on 9)

The method of claim 9, wherein the dynamic process includes a heat-up and/or cool-down process.
