This episode is about attempts to make things safer that actually make things worse. The episode focusses on the work of two specific authors, Edward Tenner (Why Things Bite Back: Technology and the Revenge of Unintended Consequences) and Lisanne Bainbridge (The Ironies of Automation). There are examples throughout the episode, but the main case studies are China Air 006 and the New Orleans Hurricane Protection System.
We’re up to 30 episodes of DisasterCast, and we still haven’t talked about the Titanic. Why start now?
This episode talks around the Titanic. We talk about icebergs, lifeboats, shipwrecks and radios, but not the sinking of the unsinkable.
The next episode will be about dangerous safety features – ways that people can or have been hurt by systems specially designed to keep them safe. If you have any suggests, post a comment to this episode, or use the feedback link above.
Safety engineering and management is full of compromises. We compromise between short term and long term risk. We compromise between absolute assurance and practicability. We compromise between blame and understanding. The one thing we can’t compromise is our professional code of conduct. Or can we? Should our actions be bound by a strict set of rules, or by their intended and expected outcomes? Is an action good because its what we would want if we were in someone elses shoes, or because it makes us into the sort of person we want to be?
And just how did McDonnell Douglas manage to stage two near-perfect dress rehearsals before a DC-10 crashed into the French woodlands?
This episode is all about level crossing safety. Level crossings are a simple situation, repeated throughout the world, that illustrate a number of important safety concepts. Through accidents such as Hixon, Lockington, Kerang, Langenweddingen and Fox River Grove we can learn lessons about human factors, dependability trade-offs, safety management interfaces and risk prioritisation.
The Introduction to System Safety course in Canberra, 7-11 April 2014. Let me know via the feedback link above if you’re coming along.
- The study of train drivers and heavy vehicle drivers mentioned in the episode
- The Hixon accident report [pdf]
- The Lockington accident report [pdf]
- The Kerang rail safety investigation report [pdf]
- The Fox River Grove Highway/RailRoad Accident Report [pdf]
In this episode we talk about Stuxnet, and the relationship between safety and security more generally.
Stuxnet demonstrated that a determined cyber attacker could influence the operation of potentially hazardous industrial machinery.
Just how representative is Stuxnet? It required elite teams from at least one, possibly two countries, and achieved only minor damage.
Are we in danger of non-state actors causing industrial accidents using their home PCs? Maybe we should ask the 14 year old who derailed
trams using a rewired TV remote control …
If you’ve ever wondered why safety is considered a systems discipline rather than simply a specialisation of chemical, civil, mechanical or electronic engineering, the humble battery is a great example. To a chemical engineer, a battery is an electrolytic process, with ions travelling from cathode to anode or vice versa, depending on what the battery is doing. To a mechanical engineer, a battery is a combined heat source and a container filled with hazardous substances. To an electronic engineer, a battery is a source or sink of electric power. In each role the battery is not a single component but is part of a variety of engineering systems. As a chemical reaction it is part of an atmospheric system. As a hot toxic container it is part of a physical layout. As a source of electricity it is part of a control or power system. For safety, we need to understand the battery as all of these things.
In this episode we’re going to discuss a number of battery hazards, along with some associated accidents.
The Feynman Gap is the gulf between engineering understanding of risk, and management understanding of risk. The concept is named after Professor Richard Feynman – drummer, lockpicker, nobel prize winning physicist and member of the Rogers Commission investigating the Challenger accident.
This episode talks about the nature of the gap, what causes it, and what we can do about it.
The accident for the episode is the Hyatt Regency Walkway Collapse of 17 July 1981. The Hyatt Regency hotel in Kansas City featured a spectacular multi-storey open atrium, crossed by suspended walkways on each floor. During a dance competition on the 17th of July, 1981, the atrium was packed with dancers and spectators. The fourth floor walkway fell onto the second floor walkway, which fell into the crowd below. At the time it was the deadliest building accident in the United States, and it still holds that unfortunate title unless you count the collapse of the South Tower of the World Trade Centre.
In the final segment I update the discussion of Bicycle safety from Episode 22 to address the topic of riding two abreast or in the middle of the lane.
This episode discusses a few aspects of preflight briefings on passenger aircraft. In particular, we look into accidents and evidence relating to lifevests, oxygen masks, and brace positions.
- Miracle on the Hudson Accident Report
- Flight ALM 980 Accident Report
- Stansted Incident Report
- Virgin America Safety Video
- Air New Zealand (Middle Earth) Safety Video
- Air New Zealand (Fit to Fly) Safety Video
- Thompson Safety Video
This episode addresses seven questions about bicycles and safety:
- How dangerous is cycling compared to walking or riding in a car?
- Does cycling actually get safer as more people cycle?
- Should cyclists wear helmets?
- Are most bicycle accidents caused by cyclists riding dangerously?
- Why is it so hard to create decent bike lanes?
- How dangerous are bicycles for pedestrians?
- What’s the deal with those cycle airbags?
- A Case-Control Study of the Effectiveness of Bicycle Safety Helmets – This is a good example of a case control study on the effectiveness of helmets. Link is to the abstract – subscription needed to access the full article.
- Nonuse of Bicycle Helmets and Risk of Fatal Head Injury – Another good example of a case control study. Full text of this one is freely available.
- Bicycle helmet Efficacy: a meta-analysis – This freely available article gives a fair summary of the overall evidence for bicycle helmets, including the neck-injury issue.
- No clear evidence from countries that have enforced the wearing of helmets – A fair summary of the impact of helmet laws on safety (freely available)
- Bike Lanes versus Wide Curb Lanes: Operational and Safety Findings and Countermeasure Recommendations [pdf] – A good example of video camera studies of cyclist behaviour and the way cycling infrastructure changes this behaviour
- Bicycle Helmet Safety Institute review of the Hovding – this is the fairest review I found of the bicycle airbag system.