1.1 Rapidly Shifting Challenges for Hardware Reliability and Software Quality
Hardware reliability and software quality, why do you need it? The major US car manufacturers saw their dominance eroded by the Japanese automobile manufacturers during the 1970s because the vehicles produced by the big three had significantly more problems. The slow downward market slide of the US automobile industry was predictable when the defect rate of US automobiles was compared with the Japanese automobile industry. In 1981, a Japaneseâmanufactured automobile averaged 240 defects per 100 cars. The US automobile manufacturers, during the same time period, were manufacturing vehicles with 280â360% more defects per 100 vehicles. General Motors averaged 670 defects per 100 cars, Ford averaged 740 defects per 100 cars, and Chrysler was the highest, with 870 defects per 100 cars.
Much has been written about how this came about and how the US manufacturers began implementing total quality management (TQM), quality circles, continuous improvement, and concurrent engineering to improve their products. Now the US automobile industry produces quality vehicles, and the perception that Japanese vehicles are better has eroded significantly. J.D. Powers and Associates reported in its 1997 model year report that cars and trucks averaged about 100 defects per 100 vehicles. This represented a 22% increase from 1996 and a 100% decrease from 1987. Vehicles such as the GM Saturn and Ford Taurus are a tribute to that success, both in financial terms and in the improved perception that automobile manufacturers in the United States can produce reliable, quality automobiles. Quality programs like TQM have dramatically improved American manufacturing quality. The automotive industry has also benefited from the quality of the components going into automobiles, which is also at a very high quality level. Counterfeit components and counterfeit material is still a major concern for the electronics and automotive industry that requires constant diligence and an effective program to minimize the risk of counterfeit material entering into the production stream.
In the 1970s, the typical automobile warranty was for 12 months or 12 000 miles. In 1997, automobile manufactures were offering 3âyear/36 000âmile bumperâtoâbumper warranties. Three years later, these same automobile manufacturers were offering 7âyear/100 000âmile warranties. Jaguar is now advertising a 7âyear/100 000âmile warranty on its used vehicles! BMW has responded with a similar type of program. The reason these manufacturers can offer longer warranty periods is because they understand why and how their vehicles are failing and can therefore produce more reliable vehicles.
A 1997 consumer reports survey of 604,000 automobile owners showed a dramatic improvement in the perception of the reliability of USâmanufactured automobiles. The improvement by the big three automobile manufacturers did not occur overnight. It was the result of a commitment to provide the necessary resources along with a credible plan for producing reliable vehicles. It was a paradigm change that took years and evolved through many steps.
The process to improve hardware reliability has made significant progress over the past 20 years. If you follow the process outlined, there can be significant improvements to your product reliability. The hardware reliability errors are often the result of either not following or poorly executing the hardware reliability process rather than being a weakness in the reliability process.
Reliability research and development continues in some areas, such as prognostics and health management (PHM). PHM can improve product maintainability, reduce unscheduled downtime, and lower the cost of ownership over the lifetime of a product. PHM uses realâtime sensors to monitor the health of a system. The sensor data is then compared to a good set of data to determine if the system is degrading and to estimate the time to failure. PHM strategies are being use in the automotive, aerospace, and other industries.
Even though hardware reliability has improved significantly for many companies, software quality and software security have become a bigger issue. This is partly due to the fact that many of the new products being developed require significantly more software and firmware. For example, McKinsey & Company estimates that over 10% of automobile vehicle content today (2018) is software and that software will reach 30% of vehicle content by 2030 [1]. Many of the hardware products being developed increasingly need software and firmware to function properly. Increasingly, software is also being deployed to manage critical safety and health operations such as robotic surgery and selfâdriving cars. It is not uncommon for the software development team to be inadequately sized for the staff and skill level needed to support software development. Automation, the Internet of things (IOT), advances in WiâFi and Bluetooth, and greater use of the internet to improve customer experience all drive the need for increased software code development and an improved software quality system. The Internet can be used to push out software updates effortlessly to the end user, but a poor software quality process results in injecting more software bugs than it fixes.
The number of softwareâtoâhardware bugs that need to be fixed during product development can be in the order of 50 to 100 : 1. The complexity of software continues to increase along with new software languages and new drivers that create compatibility issues. About 40% of software development cost is to support testing for software verification. Software quality bugs are design faults that need to be identified, prioritized, and fixed. The traditional software quality systems addressed software bugs downstream as part of the final production testing. For many of the products being developed, this is too late in the development process. This is driving the need to improve software quality and the software development process from the perspective of project management, software requirements, and performance. This includes improvements to the software development process and staffing the team with domain experts.
1.3 Competing in the Next Decade â Winners Will Compete on Reliability
The business practices of the past few decades will not be sufficient to ensure success in the twentyâfirst century. Through the years, we've learned to master the skill of building quality products. Higherâquality products have resulted in improved profit margins. In fact, consumers make buying decisions based on their perception of which products have better quality when the competing products were of the same approximate price. In the past few decades, reliability was not a deciding factor for most consumers. This is mostly the result of the consumer's lack of knowledge about product quality. However, the average consumer in the twentyâfirst century will make buying decisions based not only on price and quality but also on the perceived reliability of the product. Consumers make buying decisions based on which product offers the best value. We can define product value as
Here the customerâperceived value is related to the quality and reliability of the product. One of the key advantages of implementing reliability throughout the organization and at every phase of the product life is that the product value increases because of an improved customer perception of the value of the product and a lower cost of production. There is a common misperception that implementing reliability delays the product development time and increases the cost of the product (both in material and production costs). The reality is the exact opposite. Products that are more reliable generally have lower production costs. The reason for this is the result of many factors that contribute to reducing product costs and the product development cycle. For example, products that are reliable generally have
- Higher firstâpass yield in test,
- Less material scrap,
- Less product rework (which helps to lower product cost and improve product reliability),
- Fewer field failures,
- Reduced warranty costs (this saving can be passed onto the consumer to provide a competitive price advantage),
- Lower risk of recall,
- Superior designs that are easier to manufacture.
Looking back at the definition of what the consumer considers to be of value, it becomes clear that product reliability will increase the perceived product value and lower the cost of production. This is an important fact about product reliability that is often misunderstood.
