Click Opstakes 2016-2017 to check out this year’s Opstakes
On the walls of the main lobby of Toyota’s headquarters in Tokyo hang 3 portraits. One portrait is of the company’s founder, another is of the current chairman and the largest portrait is of Dr. Deming, the US based statistician whose work was seminal in Japan’s quality revolution.
Quality essentially is an exercise in consistency. Quality control is where we have a production process, and we want to ensure that high quality is maintained throughout the operation. We sample from the production process, and we take corrective action whenever we believe that the process is out of control- producing items that, on average, lie outside our specified target limits.
Variation and Causes:
The cause of variation in output properties can be subdivided into two groups : Assignable and Unassignable causes
Assignable Causes of variation are those identifiable reasons that can be determined. For example : improper raw material, operator mistake, jigs and fixtures errors, improper tools etc. These variations result the observation fall beyond the permissible limit.
Unassignable causes of variation are the result of randomness in the nature, sometimes called as natural variation in the process. These types of causes may include variations in the machine, fluctuations in working condition, change of state of mind of operator etc. The minimization of this type of variation requires high cost. Generally these are very small in magnitude and sometimes called as allowed variation of the process. The observations with these variations fall within permissible limit and the process is supposed to be in stable condition.
Some tools for quality control are :
Control chart is an effective tool to maintain quality. It is a graphical display of measurements of an industrial process through time, with which a quality control engineer can identify any potential problems with the production process. The idea is that when a process is in control, the variable being measured (eg the mean of every 4 observations) should remain within the stipulated “tolerance limit”.
Below is an example of a control chart
Pareto diagram is a bar chart of the various problems in production and their percentages which is used to identify the problem(s) which should be addressed first.
The basic ideology of Taguchi was that the quality of a large system deteriorates as we add the small variations in quality for all its separate components. For example, if we look at a car, the car’s quality may not be good even though all its components are within desired levels when considered alone. To solve this problem Taguchi developed the idea of a total loss to society due to the lowered quality of any given item. That loss to society is to be minimized. This is done by introducing a loss function associated with the parameter in question and by trying to create production systems that minimize this loss both for components and for finished products.
The Mumbai Dabbawalas is a 120 year old organization which solves a very critical problem for working Mumbai residents.
Mumbaikers don’t want to carry their tiffin boxes along with them on the local
1. The working Mumbaiker starts early in the morning. If he leaves home at say 7 o’ clock in the morning, he doesn’t want his wife/mother to wake up at 6 just to cook his food!
2. It’s just too crowded on the local!
Boarding a local with empty hands is difficult, with a tiffin in your hands, it becomes a mammoth task just to get on the train.
Hence come the dabbawalas who supply fresh home cooked food right to the working Mumbaiker sharp at lunch time and then carry the empty dabbas back to their homes, all at a monthly charge of approx Rs 450.
The ISO 9000 certified system are delivered 99.9999% of the time which comes at around 1 mistake per 6 million deliveries is an excellent example to demonstrate the power of standardized operational procedures. The dabbawalas have an average education level of around 8th grade and yet by following set operational procedures, the system is almost error proof with beyond six sigma quality even though there is no IT infrastructure involved.
Check out this quora article to know the entire supply chain model
Also do check out a documentary on the subject