The development of Quality in modern era could be divided into 4 phases according to D. Garvin (1988) Inspection, Statistical Quality Control, Quality Assurance and Strategic Quality Management. Each of these eras have their own characteristics and the latter is the development of the previous one. Table 1 will provide an overview on the main characteristics in each of these evolutions. Table 1. Evolution of Quality throughout time (D. Garvin, 1988)
The 19th Century marked the start of the use of Quality Control tool in industrial works, when industrialization starts to growth and the production gets centralized and its scale gets larger. The industrial revolution requires products to be manufactured identically in mass with the implementation of Fordism and Taylorism in early phase, it is hard to guarantee final products are flawless, since human error was still a huge issue then. This also mean that the old method of checking every single product which was using during the medieval time is proved to be irrelevant since it consumes too much time. Owners also starts to gain awareness on how their products quality could affect their sale numbers. So, a new method, mass inspection, is introduced, where measuring system is used by supervisors to find out if there are any unqualified products. Those products are then reworked to fit the requirements, else removed. So, the liabilities then are lying in the inspectors, not in the production managers like it used to be, which could cut down a lot of time before products are available in the market. Still, the process at that time was just basically trying to fix the issues that already occurred, not trying to proactively find the root cause of all the issues.
1922 marked a key point in the history of Quality where the first document mark formally linked inspection to quality control. In his paperwork, “The Control of Quality in Manufacturing”, G.S Radford started to recognize Quality as a standalone function from the manufacturing process. He considered “the control of quality is the correct starting point for the economy” and inspection is used as a mean to control the set-up standards of quality control (p.35). Even though the study was still in early stage, where quality control was simply about inspection, sorting, counting and grading, but Radford’s works had set up some fundamentals to modern Quality Management, like coordination between multiple departments, uniformity is the essence of quality, quality first and quantity to follow, and the early involvement of product design in manufacturing process.
Statistic Quality Control Era
Based on the first form of Taylorism, also known as scientific management, in 1931, Walter A. Shewhart put his hallmark on being the first to consider using statistics to control quality. Starting his work from mid-1920s, the statistician from Bell Laboratories views quality is “the same kind of product differ among themselves, or, in other words, the quality of a product is expected to vary”. (1931) This concept is totally new since before that time, quality is only considered after the product is finish, and to apply mathematic analysis to control quality is an innovative approach. Shewhart pointed out that there is a statistical distribution in almost all things, and by observing them it would be possible to identify the variable and stabilize the quality. He creates a quality control tool called statistical process control chart (SPC Chart) to control the variabilities of the product. This acknowledgement is the platform to establish criteria which indicate the acceptance level of quality control.
The problem with Shewhart early development is he still considering full inspection on all finished products, which at that time is proven to be time-consuming and too inefficient. So later, in 1941, two other mathematicians who also comes from Bell Laboratories, H. F. Dodge and H. G. Romig, in their paper: Single Sampling and Double Table Inspection, made a proposal that by divide the products into small lot and checking several products in each lot to see if the number of defects is acceptable or not by using probability sampling tables. This method is called acceptance sampling. Even though this method is no longer suitable for use in the world today, which will be discussed later, acceptance sampling was useful at that time where it is acceptable if there are defects that could get out.
Although these findings gain a lot of recognition throughout America, starting from 1942, but it had not been applied to other industries except for the telephone company itself. But World War II came and changed everything. The US with their policies of dealing arms and ammunition to Countries at war required weapon to be mass produced in a quick and efficient way. Bell Laboratory mathematicians are being used to set up new sampling tables for government inspectors to use. The result is the creation of the concept of acceptable quality levels (AQL). AQL is defined in ISO 2859-1 as the “quality level that is the worst tolerable”, meaning the percentage of defects to total outputs which still satisfy supplier control standards. This resolved the issue of having to use too many inspectors, and relieve the pressure on them. This practice soon spreads into other industries. Another turning point of this era is the formation of American Society for Quality Control. The organization was formed in 1946, is an aggregation of individuals and smaller societies who are enthusiasts about quality, sharing their studies and publications to each other, hosting conferences about the matter. The organization still operate until today, keeping the same mission that it has carried since the establishment, to promote the use of quality control in practice.
Up until WW II, the American is the leader in the Quality Control field, but post WW II, the Japanese started to catch up and identity some key elements on how to standardize quality. Their development was heavily influence by W. Edwards Deming, who is a student of Shewhart. He gave several lectures to engineers and top-managers concerning the control processes, notably the first lecture, where CEOs represent 80 percent of Japan capital showed up. The lectures which he gave in 1950 to the Japanese Union of Scientists and Engineers (JUSE) has helped Japan adapting beyond what Acceptance Quality Levels can do. Deming presented to them a new direction on how to tackle the Quality issue by showing them the Shewhart’s Cycles, later being referred to as Deming’s Cycles or PDSA Cycles, suggested that the Quality needs to be continuously control and improve. PDSA stands for Plan-Do-Study-Act, a series of step by steps learning method with the goal is to continually enhance a product or service.
By refining and researching the American approaches, they identify the need to tackle problems concerning Quality Control from right from the start and not until the issue has already occurred. One of the leader of these quality initiatives in Japan is Kaoru Ishikawa (石川 馨) with his development of the cause/effect diagram, also known as fishbone diagram.
To summarize, Quality Control at this point is only recognized in engineers and workers, but not on the top level. They still need to change the perception that has been around for decades. But we cannot deny that the acknowledge of the Japanese to these theories has set up the country itself to become the powerhouse in quality standards post World War II.