Blog posts on experiment

Posts selected fromManagement Blog - Engineering Blog - Investing Blog and other blogs

• Using Design of Experiments

  Design of Experiments can seem complicated but at the core it is fairly simple and powerful. By applying the proper techniques it allows you to gage the effect of several variables and, very importantly, the interactions of those variables with a small number of experiments (or tests or pilots).

  George Box is a wonderful author (and friend) who can write for mangers who are not knowledgeable about statistics and statisticians. Statistics for Discovery does a good job of explaining how organizations should use experiments to improve.

  continue reading: Using Design of Experiments
• George Box Webcast on Statistical Design in Quality Improvement

  There is great value in creating iterative processes with fast feedback to those attempting to design and improve. Box and Deming (with rapid turns of the PDSA cycle) and others promoted this 20, 30 and 40 years ago and now we get the same ideas tweaked for startups. The lean startup stuff is as closely related to Box’s ideas of experimentation as an iterative process as it is to anything else.   

  continue reading: George Box Webcast on Statistical Design in Quality Improvement
• How Google Works

  Google was driven from the beginning by engineers that sought to do what was best. Since those engineers were the founders of the company and still run the company Google has been able to keep the focus not on what is accepted as conventional wisdom but what actually works best. Google understands when you experiment things might not work out. Google’s solution is to experiement quickly and fail early (turn the pdsa cycle quickly). That is something every organization can apply.

  continue reading: How Google Works
• Making Changes and Taking Risks

  The system had to change. “We eliminated commissions, incentives, promotions, contests, P&Ls, forecasts, budgets, the entire functional organization chart,” Rodin says. It was a radical move. Contests and commissions — internal competition — were a way of life in the industry, the universal motivational tool.

  continue reading: Making Changes and Taking Risks
• Turning The PDSA Cycle Rapidly (Iteration)

  One point he made was that he often finds that organizations fail to properly “turn” the PDSA cycle (by running through it 5-15 times quickly and instead to one huge run through the PDSA cycle). One slow turn is much less effective then using it as intended to quickly test and adapt and test and adapt…

  In my experience people have difficulty articulating a theory to test (which limits the learning that can be gained). He offered a strategy to help with this: write down the key outcome that is desired. Then list the main drivers that impact that outcome. Then list design changes for each outcome to be tested with the PDSA cycle. 

  continue reading: Turning The PDSA Cycle Rapidly (Iteration)
• All Models Are Wrong But Some Are Useful

  All Models Are Wrong But Some Are Useful

  A great quote. Here is the source: George E.P. Box, Robustness in the strategy of scientific model building, page 202 of Robustness in Statistics, R.L. Launer and G.N. Wilkinson, Editors. 1979.

  See more quotes by George Box.

  [When I added this post in 2007 it was difficult to find the source online, now the source is much easier to find.]

  continue reading: All Models Are Wrong But Some Are Useful
• Illusions – Optical and Other

  Optical illusions provide a simple reminder of how easily we can think we know things that are not so.

  It is important to question what you believe; even when it is as obvious as the A square being darker than the B square. Understanding the ease with which we can reach false conclusions can be a powerful aid in improving management decision making.

  continue reading: Illusions – Optical and Other
• Factfulness – An Extremely Valuable Book

  Data is extremely valuable in helping us make decisions and evaluating the effectiveness of policy. However it is critical to be careful. It is very easy to focus on meeting targets that seem sensible – increasing the number of hospital beds – but that lead to less effective policy.

  ...

  The book relentlessly points out the great progress that has been made globally over the last 50 years and how that progress continues today and looks to be set to continue in the future. We have plenty of areas to work on improving but we should be aware of how much progress we have been making. As he points out frequently he has continually seen huge underestimation of the economic conditions in the world today. This book does a great job of presenting the real success we have achieved and the progress we can look forward to in the future.

  continue reading: Factfulness – An Extremely Valuable Book
• The New Age of Robots and What it Means for Jobs

  Industrial robots are the most advanced application of robots in business today but they are still far from plug and play solutions. They require skilled experts to have them work effectively; but the capabilities and usability have greatly increased over the last 20 years. Respect for people (and all that entails about the management system) is an important part of creating a management system to have the most success integrating robots.

  The ability of us to create technological solutions to accomplish tasks that required people has exploded in the last 20 years and will continue to. Lawyers are finding much of what they do can be done by a computer. Much, doesn’t mean all, obviously. Search and rescue in disaster areas is another task that robots are playing an increasing role in; and the use of robots will likely continue to grow quickly. Technology is taking over many aspects of medical care that were not long ago seen as requiring highly trained and experience medical professionals (reading scans, diagnosing illness…).

  
  I think many of these advances are moving so quickly that we are not properly thinking about the long term future of our organizations. The disruption these changes will take will be difficult to predict and plan for.

  ...

  continue reading: The New Age of Robots and What it Means for Jobs
• Large Scale Redox Flow Battery (700 megawatt hours)

  Scientists and engineers in Germany have created the largest battery in the world with redox flow technology.

  Redox flow batteries are liquid batteries. The Friedrich Schiller University of Jena has developed a new and forward-looking salt-free (brine) based metal-free redox flow battery. This new development will use salt caverns as energy storage.

  ...

  Both charged electrolytes can be stored for several months. The maximum storage capacity of this redox-flow battery is limited only by the size of the storage containers for the electrolyte liquids.

  The project is being ramped up now, going through a test phase before bringing the full system online; they are aiming to achieve this in 6 years. The electrical capacity of 700 megawatt hours will be enough to supply over 75,000 households with electricity for one day.

  continue reading: Large Scale Redox Flow Battery (700 megawatt hours)
• Drone Deliveries to Hospitals in Rwanda

  Partnering with the Government of Rwanda, Zipline serves 21 hospitals nation-wide. They provide instant deliveries of lifesaving blood products for 8 million Rwandans.

  Their drones are tiny airplanes (instead of the more common tiny helicopter model). Supplies are delivered using parachute drops from the drone. Landings are similar to landings on aircraft carriers (they grab a line to help slow down the drone) and, in a difference from aircraft carrier landings, the drone line drops them onto a large air cushion.

  ...

  The drones can deliver up to 50-75 km (which I believe means they must have a range for 150 km because they must return to their home base). The cost is about equivalent to the current (much slower) delivery methods (car or motor bike).

  continue reading: Drone Deliveries to Hospitals in Rwanda
• Using Design of Experiments

  Design of Experiments can seem complicated but at the core it is fairly simple and powerful. By applying the proper techniques it allows you to gage the effect of several variables and, very importantly, the interactions of those variables with a small number of experiments (or tests or pilots)...

  continue reading: Using Design of Experiments
• Science Explained: Wind Powered Vehicle Traveling Faster Than the Wind

  This is an interesting explanation of a the physics involved with vehicle propulsion. And it is a great video showing the scientific method at work.

  They only touch on it a little bit but the need for creating 4 versions of the small treadmill device to illustrate the principles in action is a great example of how science inquiry and engineering work. There are often many failed attempts before an engineering solution to the issue involved can be properly created...

  continue reading: Science Explained: Wind Powered Vehicle Traveling Faster Than the Wind
• Fooled by Randomness

  When people are asked to explain random variations in data they will make up special causes (that they often even believe are special causes even when they are not) but you can improve management a great deal by just stopping the requirement to “explain” common cause variation...

  Use that time instead to standardize processes. Create control charts for critical processes. Run experiments using PDSA cycle…

  continue reading: Fooled by Randomness
• Interview of Bill Hunter by Peter Scholtes on Statistical Variability and Interactions

  For some processes it is enough to know a couple important variables and have an understanding of how they interact to impact results. Often though problems are created because the organization doesn’t learn enough about variables that can have a substantial impact on results and therefore feels blindsided by poor results. In some of those cases they were blindsided not by unforeseeable random factors but by variables they should have learned about. And then based on that knowledge designed their processes to take into account the potential impact of variations in that variable...

  continue reading: Interview of Bill Hunter by Peter Scholtes on Statistical Variability and Interactions