International Business Machines Corporation

Michal Rosen-Zvi

Robert J Baseman

Ehud Aharoni

Oded Margalit

Noam Slonim

Adam D Ticknor

Raminderpal Singh

Hong Lin

Fateh A Tipu

Ramona Kei

Timothy M McCormack

Kevin MacKey

Marie Angelopoulos

Aisha Walcott

Shahram Ebadollahi

Bharath Dandala

Ching-Huei Tsou

William A Muth

Piyush Madan

Divya Pathak

James V Codella

Ananya Aniruddha Poddar

Ching-Hua Chen

Si Sun

Frederick A Scholl

Diwakar Mahajan

Susan G Conti

Parthasarathy Suryanarayanan

Stewart T Sill

Systems, devices, computer-implemented methods, and/or computer program products that facilitate artificial intelligence (AI)-assisted curation of non-pharmaceutical intervention (NPI) data from heterogeneous data sources. In one example, a system can comprise a processor that executes computer executable components stored in memory. The computer executable components can comprise an extraction component and a change detection component. The extraction component can extract candidate non-pharmaceutical intervention (NPI) events from data associated with a defined disease. The change detection component can evaluate the candidate NPI events for inclusion in a dataset storing NPI events in a defined format.

Artificial intelligence-assisted non-pharmaceutical intervention data curation

A system provides an intervention for a user and comprises at least one processor. The system monitors behavior and context of a user to generate a behavior history. One or more models are utilized to determine an intervention for the user to induce a behavior modification, wherein the one or more models map interventions to user context and behavior and utilize the behavior history to determine an effective intervention for the user. The intervention is provided to the user and feedback is received in response to the intervention. The one or more models are updated based on the feedback. Embodiments of the present invention further include a method and computer program product for providing an intervention to a user in substantially the same manner described above.

Personalized intervention based on machine learning of behavior change states

A method, a system and a computer program product suitable for use in a manufacturing environment comprising a multiplicity of nominally identical independent tools. A computing device generates a multi dimensional array of process trace data derived from at least one of the independent tools, wherein, the array includes data representing a first dimension comprising a list of steps in a manufacturing recipe and data representing a second dimension comprising a list of a set of sensors generating measurements from at least one of the independent tools. The computing device conducts an analysis on at least one preselected subset of the multi dimensional array for the purpose of evaluating at least one operating characteristic of at least one of the independent tools. The computing device presents results of the analysis via a set of hierarchically linked and browseable graphics.

Method and system for evaluating a machine tool operating characteristics

A method and system for evaluating a performance of a semiconductor manufacturing tool while manufacturing microelectronic devices are disclosed. At least one report is generated based on executions of at least one statistical test. The report includes at least one heat map having rows that correspond to sensors, columns that correspond to trace data obtained during recipe steps, and cells at the intersection of the rows and the columns. At least one sensor in the tool obtains trace data of a recipe step while manufacturing at least one microelectronic device. A computing device analyzes the obtained trace data to determine a level of operational significance found in the data and assigns a score to the trace data that indicates a level of operational significance. Then, the computing device places the score in a corresponding cell of the heat map. A user uses the cell for evaluating the tool performance.

Techniques for estimating a quality of one or more wafers are presented. One or more first wafers comprising one or more first dies are tested. A probability of wafer failure is determined in accordance with one or more first test measurements of the one or more first dies. A pass status and/or a fail status of one or more second wafers is inferred by testing a select one or more second dies of the one or more second wafers and evaluating one or more second test measurements of the select one or more second dies in accordance with the determined probability of wafer failure.

Predicting wafer failure using learned probability

Growing community of inventors

Jerusalem, Israel

Michal Rosen-Zvi