Semiconductor reliability using random and wearout failure models

Flexible manufacturing cells (FMC) are used to process a variety of products in the system and provide higher productivity. The components of the system such as robots and machines are more likely to fail when compared with the traditional manufacturing system due to their higher operating rate. Literature studied reveals that the failures of the machines have a high impact on the production rate of the system. In this work, two types of failures are considered i.e. wear-out failure and random failure. Wear-out failure can be eliminated by applying some maintenance plans but random failures are difficult to eliminate. So simulation experiments along with maintenance plans are studied to eliminate such failures. We also discussed two types of failure rate distributions i.e. exponential, and Weibull distribution and their effect on the throughput of the FMC. In this analysis, ARENA simulation model is made and simulation experiments were performed to analyze the throughput of FMC that consists of 2 machines in parallel and a robot for material handling. We assumed that failure is occurring in machines only and the robot is reliable throughout the simulation.

Semiconductor manufacturing is mainly characterized by diversity of products, different process types, and random failures. It is extremely difficult to solve scheduling problems due to high frequency of disturbance occurrence. This paper proposes a new approach for the online manufacturing rescheduling method. Instead of having a dispatching process based on dispatching rules once again, we revise the existing schedule based on message passing principle once the disturbance occurs, and improve the performance of the revised schedule by the introduction of a quick local search according to permutations of processes on semi-critical paths. The proposed method can release the influence of disturbances in less than 1 second, and finish rescheduling process in less than 1 minute. Through the actual problems with about 200,000 processes, this method is evaluated to be more effective for actual manufacturing than the dispatching-rule based methods, some of which have been applied in many facilities

In this study, medical equipment life by bath-tab curve is analyzed, and the validity and reliability of the maintenance inventory of medical equipment is examined. A total of 7, 030 repair reports were extracted from maintenance inventory of medical equipment between 1988 and 2003, as records for the study. An indication of early failure period was mean time to first failure (MTTFF). An indication of random failure periods for exponential distribution were mean time between failures (MTBF) and MTBF (low), The internal consistency for medical equipment was analyzed using X^2 confidence level 90%. An indication of wear-out failure periods for normal distribution were mean time to repair (MTTR), steady availability, and maintenance cost (average, standard deviation). MTTFF of early failure period was 1.5 year. MTBF and MTBF (low) of random failure periods were a year, and half year, respectively. Maintenance cost, MTTR and steady availability of wear-out failure periods were 120, 000 yen, 9.5 day, and 100.5%, respectively. The result indicated the effectiveness of preventive maintenance for medical equipment. This study shows that the wear-out failure, random failure and early failure periods of medical equipment life by bath-tab curve.

The reliability of circular electrical connectors was usually assessed according to standards like MIL-HDBK-217 (Reliability Prediction of Electronic Equipment). Given to their limitations and mislead results, a new assessment method needs to be presented. Some concerns on the physics of failure (PoF) modeling and reliability assessment of circular electrical connectors are addressed in this paper to overcome this defect. An original approach combining laboratory accelerated degradation testing (ADT) and PoF modeling is proposed for effectively assessing connector reliability. Random vibration and current stress are selected as acceleration factors that affect the contact life of electrical connectors. Then, an appropriate multiple-stresses ADT scheme is derived. The PoF model for electrical connectors under multiple stresses yielded the generalized Eyring relationship. According to the system reliability model for multiposition connectors, connectors' life follows Weibull distribution, which is derived from extreme-value distribution theory. Through ADT and data gained from tests, statistical analysis result yielded the estimated value of reliability character of MIL-C-38999 I series electrical connectors under the action of vibration and current stresses.

Modern products are complicated and standards of reliability are higher than ever before. Therefore, prediction of system reliability is essential, especially as the time available for developing products has decreased. Most of the previous studies assumed that the causes of failure are random. That is, the failure rates of components follow exponential distributions. To reflect various types of failure such as initial, random, or wear-out failure, we developed a novel procedure for estimating system reliability based on the Weibull distribution. We analyzed the effect of component selection on the types of system failure. Finally, we verified the proposed method by analyzing warranty data on actual Korean products.

This paper presents a new approach for monitoring and estimating device reliability of nanometer-scale devices prior to fabrication. A four-layer architecture exhibiting a large immunity to permanent as well as random failures is used. A complete Monte Carlo based tool for a-priori functional fault tolerance analysis was developed, that induces different failure models, and does subsequent evaluation of system reliability under realistic constraints. A structured fault modeling architecture is also proposed, which is together with the tool a part of the new reliability design method representing a compatible improvement of existing IC design methodologies

The semiconductor manufacturing is one of the most sophisticated processes in modern industries. It contains hundreds of complex operations, and this makes semiconductor reliability a challenging task especially in recent years that integrated circuits (ICs) have been broadly used at much wider areas including the health cares, automotive, and aerospace applications, where safety is the major concern. To make things worse, due to the shorter product life cycles and severe competitions, time-consuming reliability tests become unacceptable. The industries are in urgent need of a fast and effective way to quickly assess reliability risks. Under all these stringent limitations and the high reliability targets, we propose a consolidated approach that consists of an efficient and effective reliability baseline (BL) management system and a test scheme developed based on sequential probability ratio test (SPRT). To ensure a trustworthy setting for the SPRT test scheme, we apply the Bayesian approach to continuously update the BL database. An integrated flow chart is introduced to elucidate our proposed methodology. It is the first time SPRT is applied on reliability risk assessments with two unknown parameters. Several actual cases from semiconductor arena are reported to demonstrate the merits and powers of our proposed methodology. From these real cases, as a side benefit, we show the wide applicability of our method, which does not limit to reliability but on yield and functionality checks as well. To facilitate the use of our method by field practitioners, we do the best to explain derivations in a plain manner and provide analysis codes in R (in Supplementary Material) to make our approach as portable as possible.

Combining with the practical engineering case, this paper decomposed and analyzed the process of electronic products’ reliability assessment based on the Physics of Failure (PoF) method. The first step is to build a digital prototype model and then a PoF model. The second step is to use the Monte Carlo sampling method to obtain the failure samples of different fault points based on the PoF model. The third step is forming the failure DFs by fitting the failure samples. The fourth step is using the failure DF to calculate the Time to Failure (TTF)of the components. Finally, the paper can obtain the time of the TTF and the failure DF by applying the competitive failure principle. Compared with the reliability evaluation method based on test data, the reliability evaluation method based on PoF is more economical and efficient.

The aim of this study is to explore the relationship between external threats and creativity in China’s technology industry through the lens of U.S.–China semiconductor industry frictions. Since the 2010s, the U.S. has progressively intensified its pressure on China’s semiconductor industry, leading to fierce tensions between the two nations. Both countries view the semiconductor industry as critical to national security, and their disputes over semiconductors are framed as security issues. This study integrates the theory of “Creative Insecurity” and optimizes it with empirical adjustments according to the characteristics of the semiconductor industry. By employing a mixed-methods approach that combines qualitative and quantitative analysis, the study examines U.S. sanctions policies during the Obama, Trump, and Biden administrations, along with China’s policy responses and the development of its semiconductor industry. The study establishes correlations between the intensity of external threats from the U.S. and China’s semiconductor-related metrics, such as technological patent levels, process advancement, and self-sufficiency. The findings conclude that U.S. sanctions on China’s semiconductor industry are positively correlated with its development. Sanctions from the United States may stimulate the growth of China’s independent semiconductor industry.

This paper discusses educational tools, resources and experience in teaching physics-of-failure (PoF) modeling and data analysis methods used in accelerated life testing (ALT) and accelerated degradation testing (ADT). The concepts introduced in this paper include several probabilistic and highly technical approaches such as the PoF and mechanistic approaches to reliability prediction and assessments as well as advanced Bayesian analysis methods used in ALT and ADT to analyze life data obtained through various methods of constant and variable stress testing efforts. Other resources include codes and algorithms from the readily available computational tools R, MATLAB, and WinBUGS for solving advanced model and life estimation methods related to PoF, ALT and ADT analyses. This paper provides an overview of these resources and several example problems.

This chapter will review the historical product recalls including natural hazard and the methodology of its reliability assessment that were developed in the last century. Based on product specifications, engineer would develop new mechanism and its structure. In marketplace product recalls frequently happen. They come from the inheritance design defects in the problematic parts and are determined by the lifetime of product. If product is subjected to repeated loads or overloading and there is faulty design, product failure suddenly arises in its lifetime. To prevent it, engineers in the previous century have developed new reliability concepts such as the bathtub curve, Weibull analysis, data analysis, and the others. For instance, the frequent derail accidents of railroad in the early of nineteen century started the research for its root cause and made the S-N Curve. The chronic failed vacuum tube in the WWII created the bathtub curve. As NASA developed for the space shuttle program in the mid-sixties, FMEA, FTA and Weibull analysis for reliability testing today have been widely used in company. Now since Integrated Circuit (IC), transistor radio and TV in the late of 1960s are introduced, Physics Of Failure (POF) become more important tools to analyze the failure mechanics in product. However, in the field of mechanical/civil system, representative POFs were still fracture and fatigue. As improperly choosing shape and material in the design process, product has faulty design—enough strength and stiffness in the final structure of product. As a solution mechanical engineer should find the problematic parts by reliability testing method and modify them before product launches in market.

Due to the recent advances in technology, the systems are becoming more demanding in terms of functionality and implementation complexity. Therefore, when system failures are involved in such complex systems, the effects of the related safety issues can also be more serious, thereby causing in the worst case irrecoverable hazards on both human being and properties. This fact can be witnessed in the recent rail systems accidents. In general, the accidents can be attributed to the systematic failure or the random failure. The latter is due to the aging or unsatisfied quality of the parts used in implementation or some unexpected external cause that would otherwise result in accidents whereas the former is usually related to incomplete systems design. As the systems are becoming more complex, so are the systematic failures. The objective of the paper is to study an approach to solving the systematic failure. To do so, at first the system design process is augmented by the functional safety activities that are suggested in the standard IEC 61508. Analyzing the artifacts of the integrated process yields the traceability, which satisfies the requirements for reduction of systematic failure as provided in ISO 26262. In order to reduce systematic failure, the results are utilized in the conceptual design stage of systems development in which systems requirements are generated and functional architecture is developed.

In this paper, we present a methodology to derive component failure rate and failure mode data for mechanical components used in automation systems based on warranty and field failure data as well as expert opinion. We describe a process for incorporating new component information into the database as it becomes available. The method emphasizes random mechanical component failures of importance in the world of safety analysis as opposed to the wear-out and aging mechanical failures that have dominated mechanical reliability analysis. The method provides a level of accuracy significantly better than warranty failure data analysis alone. The derived database has the same form as that for electrical/electronics databases used in FMEDA analyses used to show compliance with international performance-based safety standards. Thus, the mechanical database can be used in conjunction with existing electrical/electronics databases to perform required probabilistic safety analysis on automation systems comprised of both electrical and mechanical components.

Traditional reliability prediction methods are being confounded by current and near future semiconductor technologies, as gate feature sizes shrink below 100 nanometers (nm) causing the emergence of atomic level failure mechanisms and early wearout. These devices and their failure characteristics are rapidly changing as the semiconductor industry aggressively pursues scaling in a highly competitive marketplace. The Physics of Failure (PoF) approach to reliability has advantages for assessing these technologies. Industry groups are adapting PoF research results for use in predicting reliability for these technologies. This paper describes industry collaborative efforts in developing new reliability prediction approaches to meet future industry challenges.

Review of shelf life evaluation methods and a physics of failure approach for shelf life estimation for electronic components