
Nova Ltd
About the company
A global technology company specializing in advanced metrology and process control solutions for the semiconductor manufacturing industry. Headquartered in Rehovot, Israel, Nova was founded in 1993, then known as Nova Measuring Instruments, and currently listed on NASDAQ and the Tel Aviv Stock Exchange as NVMI.
Nova’s core expertise lies in dimensional, materials, and chemical metrology—critical technologies that enable semiconductor manufacturers to measure, analyze, and control complex fabrication processes with extreme precision. The company pioneered the concept of integrated metrology, embedding measurement capabilities directly into production tools to provide real-time, wafer-to-wafer process insight. Its portfolio combines high precision hardware with advanced modeling, software, machine learning, and big data analytics to improve yields, accelerate time to market, and support next generation semiconductor nodes.
Nova’s solutions are deployed across major front end and back end manufacturing steps, as well as in Advanced Packaging. The company serves leading logic, memory, foundry, and packaging manufacturers worldwide, supported by R&D centers in Israel, Germany, and the United States, and customer-facing operations across Asia, Europe, and North America.
The mission of Nova is to “redefine process insight” by enabling customers to see and control what was previously unmeasurable in semiconductor production. Through continuous innovation and close collaboration with top-tier chipmakers, Nova holds a strong competitive position as a leading independent provider of advanced metrology solutions in the global semiconductor equipment market, with consistent year-by-year record breaking growth and expansion.
Project contribution
Nova is WP3 leader and task leader, with three main objectives:
Holistic metrology channel optimization algorithm: to enable the determination of the minimal subset of metrology channels required to solve the costumer’s parameter of interest, per a given target and a given accuracy-level requirement.
Holistic extrapolation utility: simultaneously solve metrology targets across different test sites, or different wafers, or possibly different lots, based on known similarities between the said targets.
Hybridize physical-modeling with machine-learning: achieve complementarity such that the hybrid algo will add result interpretability, possibly leading to root-cause-analysis, all while maximizing the available data’s utility.