Manufacturing AUTOMATION

Features Distorted Realities Opinion
Designing for safety: Don’t forget about the invisible hand


Safety considerations remind me of the early days of quality. Years ago, product manufacturers tried to improve quality by inspection, and not by redesign. Manufacturers would rely on the recommendations of high-priced consultants to sort the good products from the bad. This practice did not increase the intrinsic quality of the product, nor do we improve intrinsic reliability with slogans, stickers and inspection.

Safety must be considered at product and process inception, and not as an after-thought. It should also be considered in the environment of the application. Safety is not the panacea or a silver bullet; it is more of a statistical reduction in danger.

In thinking about protecting both carbon life and profit, we must make safety considerations transparent to the application. No user training should be required. I called a buddy in Chicago and asked about this issue. He told me, “Explosion-proof cabinets are not secured after maintenance.” In other words, the cabinets are not explosion-proof if they are left unsecured. These cabinets should be designed for maintenance access and still be explosion-proof. Latches and operation must be interlocked with safety systems.

Good examples of transparent safety designs include: Daylight headlights; ABS brakes; LED tail lights; auto off coffee makers; airbags; shoes with steel toe protection; stairs with lower risers and larger platforms; heaters with tip-over protection; GFI electric outlets; and water misting at Nascar entrances. These are examples of products that statistically protect people by using the invisible hand of good conceptual design.

Advertisment

We can increase plant safety by using adequate lighting, and painting dangerous projections bright orange. Concentrated energy sources need the invisible hand as well. Fuel tanks, electric power and batteries can all be made secure. The auto racing bladder fuel tanks with auto shut down should serve as an implementation for consumer automobiles and motorcycles. Laptop batteries could be protected by a high temperature auto shutoff in the battery. This costs less than a total recall and redesign. No design is perfect, but we can sharply lower the likelihood of batteries overheating.

Even restroom infections are being reduced by auto flushing controls and using alcohol creams after washing.

I’ll share one of my favourite stories. I was on the road for a week in Chicago at a manufacturing gig. Just before I left for Chicago, a new ward was accepted by my noble bride, Shirley. Soon after I arrived in the windy city, Shirley called. The new kid did not know how to get ESPN on the living room TV. The remote runs all the gadgets and the satellite system. There must be a hundred keys on the two remotes. I tried to communicate the control instructions over the phone but with little success. I surrendered, and told our visitor to just hack it. The next day I learned that he successfully got ESPN on TV. When I came home from the trip and tried to watch TV, all of the commands were in Spanish. It took me an hour to figure out how to return the commands to English.

What does this have to do with safety? Safety needs to be silently communicated to the user. Communication needs to be obvious and designs should take into account the next untrained user on the system.

We need to learn from our mistakes. Black boxes on aircrafts are useful devices to verify that errors do not repeat. Perhaps we need a black box for our factories, software and processes. I contend that putting liability warnings on a chain saw does not make the saw less safe. Training surely helps, but is not the complete answer. Removing the “safety defect” is a large part of the solution. Copper wires can be broken by cranes. Wireless factories are safer since error rates will not be caused by physical trauma.

Long ago, I worked on MUMPS, a clinical software package for hospitals. I was concerned about error rates in the software. My boss asked, “Do you know the error rate for people systems?” Capitulation is now my middle name. The VA medical system went from one of the worst health systems to one of the best by using modern technology.

Do not be afraid! Modern technology can make the world safer for all when correctly applied.


Dick Morley is the inventor of the PLC, an author, speaker, automation industry maverick and a self-proclaimed ubergeek. E-mail him at morley@barn.org.