[-AI-3.5]Can quality theory using statistical methods survive in the 21st century?

Abstract

生産ラインにおける品質管理に実装された統計的手法は、過去数十年間生産性の向上に貢献してきました。しかし、現代の生産環境では、適用される統計解析モデルがますます広範な事例に対応できなくなっています。特に、正規分布モデルに基づく品質管理ツールは、複雑な現代生産プロセスに適用するのがますます困難になっています。生産管理者は品質理論を再検討し、革新的な手法を導入する必要があるかもしれません。次の章では、この必要性について詳しく調べてみます。「品質理論の進化」をテーマに、品質管理の歴史的変化を振り返ると、ウォルター・A・シューハートによって1920年代に提案された統計的品質管理（S.Q.C）が広まり、品質の統計的アプローチが広がりました。そして、1960年代には、日本の自動車産業において統計的品質管理（S.Q.M）が品質管理方法として確立され、すぐに注目されました。21世紀における統計分析の必要性を考えると、生産現場の複雑さが増し、より高次元で精密な品質管理が求められるようになり、その重要性が増しています。

Table of contents.

• Introduction
• Evolution of Quality Theory
• Current State of Statistical Methods
• Limitations and Issues
• Emergence of Innovative Methods
• Necessity of Quality Theory in Modern Times
• Conclusion

-- CAUTION

[Text]This blog is purely generated by Cohesive.ai

Introduction

The statistical methods implemented in quality management on the production line have contributed to productivity improvements for the past few decades. However, in modern production environments, the applied statistical analysis models are becoming increasingly unable to cope with a broad range of cases. In particular, quality management tools based on normal distribution model are becoming more difficult to apply to complex modern production processes. Production managers may need to revisit quality theory and introduce innovative methods. In the following chapter, we will delve deeper into this need.

Evolution of Quality Theory

Looking back at the historical changes in quality control, Statistical Quality Control (S.Q.C) was proposed by Walter A. Shewhart in the 1920s, and the statistical approach to quality then spread. Then, in the 1960s, Statistical Quality Management (S.Q.M) was established as quality management method in Japan’s automobile industry, and it quickly caught attention.

Considering the necessity of statistical analysis in the 21st century, the complexity of production sites has increased, and higher-level and more precise quality management is required, increasing its importance. However, statistical methods alone cannot solve everything.

The current state of statistical methods.

In modern production sites, issues have been pointed out regarding the conventional statistical methods for quality control. Let’s consider the factors for this below.

Regarding statistical methods up until now, it can be said that methods based on normal distribution model were mainstream. However, in actual production sites, there are many non-normal distribution data caused by equipment and human factors, which do not follow normal distribution. As a result, it can be said that it is necessary to shift towards methods based on non-normal distribution models in the future, as conventional methods are difficult to deal with in this regard.

In addition, in modern production sites, complexity has increased in evaluating the quality of products due to the expansion of product variations and diversification of parts. Therefore, problems have arisen that cannot be fully addressed by conventional methods.

From the above, it can be said that the appearance of innovative methods is essential for quality control.

Limitations and issues

There are many limitations to normal distribution model currently used until today. For example, predictions become impossible for events with very low probabilities, and extreme events may have higher estimated probabilities. However, precise statistical management using normal distribution model has been proven necessary by many companies to maintain consistent product quality.

On the other hand, there are many issues present in modern production lines. For instance, production processes have become complicated, and management of diversified product lines is required. Moreover, there has been an increase in cases where quality of raw materials is inconsistent. Hence, statistical analysis using methods that assume non-normal distributions has become necessary. Therefore, more innovative methods are required to maintain product quality.

The emergence of innovative techniques

In modern production sites, there are many issues that cannot be solved by the normal distribution model. As a result, non-statistical methods based on reformed defect detection and planned actions have emerged. This enables the resolution of problems that couldn’t be found with previous methods.

There are various methods for improving defect detection, for example, introducing detection systems that use AI, identifying improvement points based on past problems and training employees. With these methods, it is possible to reduce the occurrence rate of defective products.

Non-statistical methods based on planned actions focus on identifying the causes of defective product occurrence. By using this method, the efficiency of the production line can be improved. For example, by minimising the time taken to exchange information in the production line, productivity can be increased.

In order to achieve more efficient production lines, these new methods are crucial. It’s impossible to respond to the problems that occur in modern production lines using only statistical methods. By adopting new methods, productivity can be improved and competitiveness maintained.

The necessity of quality theory in modern times

Production sites are becoming increasingly complex. This complexity presents a major challenge for quality management. Compared to the past, quality management needs to be more delicate and precise. In modern production sites, a variety of materials and complex processes overlap. Therefore, it is natural for quality management to occupy an important position. Regulations and requirements regarding quality are becoming increasingly stringent, making quality management an increasingly important issue for organisations. If quality management is not sufficient, it can lead to significant economic losses, such as product defects, recalls, and reduced brand value. Therefore, it is expected that quality theory will continue to evolve in the future. Improving production efficiency is also an important issue for quality management. If production efficiency is low, production costs rise, and product prices increase. Therefore, quality management in production sites requires statistical and non-statistical approaches, such as improvement activities and action plans. Therefore, to achieve accurate quality management and production efficiency in modern production sites, it is essential to adopt new technologies and techniques when quality theory becomes outdated or ineffective.

Conclusion

In modern production sites, the limitations of quality theories have become apparent. Not only statistical methods, but also innovative approaches need to be adopted to cope with more complex production sites.