The user is now an integral part of the evaluation process that determines the safety and reliability for a variety of products. To enhance a product’s safety and reliability, a company can go through several possible evaluation steps that may include:
• Design reviews to verify the design and selection of components.
• Incoming inspection to check components used in the product.
• Testing at the subsystem and system level that simulates actual product usage.
• Quality control to inspect the product’s workmanship.
- UL Battery Safety Standards now FDA Recognized Consensus Standards for Medical Devices
- 1 W DC-DC Converters Support cULus 60950 Safety Certification
- Circular Connectors Address Strict Safety Requirements For Lightning Protection
- Series of SMD MLCCs for Safety-Certified Applications
- Intelligent, Universal Safety Relay With Reduced Footprint
Regardless of the test and inspection results, which can minimize future product problems, the user is the ultimate judge of a product’s performance. User experience provides the only real safety and reliability data fed back to the manufacturer and government agencies that can initiate recalls and repairs.
Automobiles are a good example of user participationin evaluating product performance. User input has led to the recall list for vehicles identified by the National Highway Traffic Safety Administration (NHTSA). Fig. 1 is an Audi model that was recalled.
The most recent list includes:
• 442,000 Volkswagen vehicles from 2011-2013, for the trailing arm that may have reduced durability, causing a loss of vehicle control.
• 28,037 Range Rover vehicles from 2013-2014, for the Tire Pressure Monitoring System (TPMS) that may fail to locate the sensor, increasing the risk of tire failure.
• 5,805 MINI Cooper Hardtop vehicles from 2014, for the spare wheel that may separate from the vehicle while driving.
• 393 Volkswagen Jetta and Passat vehicles from 2015, for potential unexpected movement of the driver’s seatback due to a bracket which may not engage correctly.
• 1,509 Mercedes-Benz C300 and C400 4Matic vehicles from 2015, for the sliding shaft that may disengage from the steering column, resulting in a loss of steering control.
• 52,738 Nissan vehicles from 2003-2006, for the air bag inflator that may rupture upon deployment of the air bags.
• Honda vehicles from 2002-2006, for the air bag inflator that may rupture upon deployment of the air bags.
• 6,562 Infiniti vehicles from 2013-2015, including Q70 and M35 hybrid models, for a potential software error that may cause unexpected acceleration.
• 5,412 Infiniti hybrid vehicles from 2014, including Q50 and Q70 models, for a software error that may cause the electric motor to stop working.
• 817 Infiniti hybrid vehicles from 2014, including Q50 and Q70 models, for potentially damaged transmission housings that could crack and disable the vehicle.
• 20,000 Toyota vehicles from 2014-2015, including Toyota Sienna and Lexus RX350 models, for a potential fuel leak, increasing the risk of a fire.
1,848 Infiniti QX56 and QX80 vehicles from 2013-2014, for the air bag inflator that may have been manufactured with an incorrect part, causing the inflator to rupture.
Automobiles aren’t the only products that have been subjected to recalls. Consumer products have also been recalled in response to faults found by the user. A recent list of product recalls listed by the Consumer Product Safety Commission (CPSC) includes:
- Tyco Fire Protection, for Simplex Fire Alarm Control panels that fail to activate.
- Bob-Cat, for crash hazards of zero turn mowers
- Leatherman, for laceration hazard in a children’s multi-tool
- Graco, for fingertip amputation hazard of 11 models of strollers (Fig. 2)
- Hoppe’s, for failure to meet child-resistant closure requirements in a semi-auto bore cleaner
- Mohawk, for a rug’s fire hazard.
- AP Specialties, for the fire hazard of power bank chargers.
There also have been medicine recalls by the Food and Drug Administration (FDA) because there was a reasonable probability that the use of or exposure to a product might cause serious adverse health consequences or even death. In many cases user inputs prompted the company to issue its own voluntary recall. Recent recalls by the FDA included:
The user has also been active in discovering software “bugs.” It is hard to tell who is credited with using the word “bug” first. However, Thomas Edison used the word in an excerpt from a letter he wrote in 1878 to Theodore Puskas, as cited in The Yale Book of Quotations (2006): “'Bugs' -- as such little faults and difficulties are called -- show themselves and months of intense watching, study and labor are requisite before commercial success or failure is certainly reached.”
Software testing for the pursuit of bugs is an investigation conducted to provide stakeholders with information about the quality of the product or service under test. Software testing can also provide an objective, independent view of the software to allow the business to appreciate and understand the risks of software implementation.
In general, properties that indicate the extent to which a system under test meets the requirements include:
• Responds correctly to all kinds of inputs.
• Performs its functions within an acceptable time.
• Is sufficiently usable.
• Can be installed and run in its intended environments.
• Achieves the general result its stakeholders desire.
As the number of possible tests for even simple software components is practically infinite, all software testing uses some strategy to select tests that are feasible for the available time and resources. As a result, software testing typically (but not exclusively) attempts to execute a program or application with the intent of finding software bugs.
Users have informed the associated company of software errors that have affected them and required repairs. BCW (Business Computing World) provided examples of some recent software errors and the resulting consequences.
1. A software error in the investment model used to manage client assets resulted in an international financial services giant being fined $25 million by the US Securities and Exchange Commission (SEC). The company also had to repay the $217 million backers lost when told that market volatility rather than software failure was to blame for their investment losses.
2. Computer system problems at one of Japan’s largest banks resulted in a nationwide ATM network of more than 5,600 machines going offline for 24 hours, internet banking services being shut down for three days, delays in salary payments worth $1.5 billion into the accounts of 620,000 people and a backlog of more than 1 million unprocessed payments worth around $9 billion.
3. An Australian bank began giving out large sums of money from 40 cash machines across one city. Officials at the company said they were operating in stand-by mode, so could not identify the account balances of customers.
4. Core and back-up switch failures resulted in network services across Europe, the Middle East, Africa and Latin America going down for 3-4 days. The blackout left millions without email, web browsing or instant messaging services and were reportedly due to server problems at a data center.
5. Just hours after its release, a social networking site’s long-awaited tablet app was already receiving reports about minor bugs from clicking through to pages via panel icons to problems posting comments.
6. A new NZ $54.5 million computer system linking New South Wales courts and allowing documents to be lodged electronically led to damages claims for unlawful arrest and malicious prosecution, after 3,600 defects in the electronic transfer of data from the courts to the police’s database led to the wrongful arrest of 22 individuals.
7. A glitch in the automaker‘s software design and testing approach meant airbags may not be deployed for passengers in the right rear seat during a crash. This resulted in the recall of 47,401 vehicles in the US and a further 3,099 in Canada and Mexico.
8. A Japanese car company was forced to initiate a worldwide recall of over one million vehicles affected by a design flaw allowing residue from window cleaners to accumulate, which can degrade the switch’s electrical contacts and potentially cause a fire over time. This recall followed a global 2.5 million recall by the same company due to design flaws that allowed vehicles to shift out of park and engine stalls.
9. After a software glitch that resulted in customers hitting their data limits early, some 47,000 customers, who were overcharged, but were reimbursed by a New Zealand telecom company in a NZ $2.7 million payout.
10. An army computing system designed to share real-time intelligence with troops on the front line has hindered troops by being unable to perform simple analytical tasks. The $2.7 billion cloud-based computing network system runs slowly when multiple users are on the system at the same time and the system’s search tool made finding the reports difficult as the information mapping software was not compatible with the army’s existing search software.
More Testing Problems
Even the U.S. education system has been affected by software problems. As reported in the Los Angeles Times, the Los Angeles Unified School District's student information system, which cost more than $130 million, has become a technological disaster. The system made its debut this semester and promptly overloaded the district's database servers, requiring an emergency re-engineering. In the days and weeks that followed, many teachers were unable to enter grades or attendance or even figure out which students were enrolled in class.
Because of scheduling blunders partly stemming from the new system, students at Jefferson High School sat in the auditorium for weeks waiting to be assigned classes.
More than 600 fixes or enhancements are needed in the software, and there are "data quality and integrity issues" that include grades, assignments and even students disappearing from the system. The school superintendent said it could take a year to work out kinks in the system just to enter grades.
The question related to all these software problems is why weren’t the systems tested before they were implemented? There must be a logical way to test these systems. Perhaps there were time constraints to finish designing the software, but without comprehensive testing more time is required to fix the software. There is an old saying that is probably true: “there is never enough time to get the job done right, but plenty of time to fix it afterward.”