Hybrid quantum annealing genetic algorithm with auxiliary resource dispatching for TFT-LCD array photolithography scheduling and an empirical study

Abstract

Photolithography that relies on capital-intensive scanners or steppers has been the bottleneck of thin-film transistor-liquid crystal display (TFT-LCD) and semiconductor manufacturing. Photomask as the critical auxiliary resource for photolithography may affect the effectiveness of the TFT-LCD array scheduling. Indeed, as photomasks are increasingly expensive, it is crucial to integrate photolithography scheduling and photomask dispatching that will require managing larger substrates, longer exposure times, and mask-sharing across multiple panel designs among different steppers, complicating effective allocation of photomasks and related auxiliary resources for optimizing the photolithography productivity and overall production efficiency. However, little research has been done to address the present problem for TFT-LCD array photolithography scheduling as a whole and develop effective dispatching solutions for crucial auxiliary resources. Focusing on the realistic needs of TFT-LCD array production, this study aims to develop a hybrid quantum annealing genetic algorithm (HQAGA) integrating the dispatch of auxiliary resources for effective photolithography scheduling. An empirical study was conducted in a leading TFT-LCD array fabrication facility for validation. Comparing to the existing practice and conventional approaches, the results have shown the practical viability of the developed solution, in which the proposed quantum annealing mutation mechanism can obtain desirable solutions of explicit schedules in limited computing time. Indeed, the developed solution is implemented in this case company.