Monthly Archives: February 2009

can operations research lead the way in “combinatorial innovation”

Google’s chief economist, Hal Varian (professor of information sciences, business, and economics at the University of California at Berkeley), claims that we are in a period of “combinatorial innovation,” where understanding how to leverage technology to understand data and make informed decisions is necessary for success.  He says:

Now what we see is a period where you have Internet components, where you have software, protocols, languages, and capabilities to combine these component parts in ways that create totally new innovations. The great thing about the current period is that component parts are all bits. That means you never run out of them. You can reproduce them, you can duplicate them, you can spread them around the world, and you can have thousands and tens of thousands of innovators combining or recombining the same component parts to create new innovation. So there’s no shortage. There are no inventory delays. It’s a situation where the components are available for everyone, and so we get this tremendous burst of innovation that we’re seeing… I keep saying the sexy job in the next ten years will be statisticians. [emphasis added] Link to article.

Sure, statistics is great, but it also sounds like OR is needed!

On a related note, I read a story today that made me feel warm and fuzzy inside.  Many tech companies are trying to lure back women in science, engineering, and technology who have left the field (such as stay at home mothers).   Many companies are offering re-entry programs for women, despite the economic climate.  The article states:

While 41% of highly qualified scientists, engineers and technicians in lower-tier jobs are female, more than half eventually quit midcareer, based on research by the Center for Work-Life Policy’s Sylvia Hewlett and others, published last year in the Harvard Business Review. Women in these fields face isolation, extreme job pressures and long hours; they often become most discouraged about 10 years into their careers — just as family pressures also tend to intensify.  Still, after years at home, many women scientists and engineers yearn to return to research and development.  Link to article.

This is great news and contrary to what I have heard (namely, that once a woman leaves science/engineering/technology, it’s almost impossible to return).  Yeah!

Related posts:


Oscar “winners”

This excellent WSJ article by Carl Bialik examines the voting system in the Oscars. The Oscar voting system uses instant runoff (as opposed to the plurality system used in most US elections).

The nominees are selected using a system called instant runoff, which has been adopted in some municipal and state elections. Out of last year’s 281 eligible films, each voter selects five nominees in order of preference for, say, best picture. All movies without any first-place votes are eliminated. One problem with that system is a kind of squeaky-wheel phenomenon: A movie that is second place on every ballot will lose out to one that ranks first on only 20% of ballots but is hated by everyone else. Then, in another upside-down outcome, a movie can win for best picture even if 79% of voters hated it so long as they split their votes evenly among the losing films.

The full article illustrates this issue with a few examples that illustrate the voting problems well.

Rumor has it that the Oscar voting method prevented the masterpiece Hoop Dreams from receiving a Best Documentary nomination in 1994 (This is a travesty–in my opinion Hoop Dreams is the holy grail of documentaries).

I went to a seminar on problems in different voting systems before the 2000 elections. I wish I remember who gave the seminar, because it was one of the best I have ever attended. Ultimately, circular reasoning takes over (whoever wins the election is who we want and we determine who we want by holding an election), but there are some clear examples of who people want losing in elections. Read more about some of these issues on the Numbers Guy blog and check out this voting expert Steven Brams interview.

Who are you rooting for in the Oscars tonight?  I confess, I have not seen any of the Best Picture nominees.


Spring 2009 Conferences, Symposiums, and Workshops

I often do not hear about many small or specialized conferences, symposiums, and workshops. I would like to put together a list of upcoming meetings of interested to the OR crowd. If you know of such a meeting (conference, etc.), please post information in the comments section. The meeting could be in any discipline, so long as you think it has a sizable OR bent. Also, please indicate if there is any travel funding for students. Don’t be shy–I will approve anonymous comments.

I need your help to make this work. I’ll get things started with a colloquium that a reader told me about.

Please list meetings only for the Spring 2009 (now to May). I’ll start a post about summer 2009 meetings soon. To find these posts, I created a category (“Conferences, Symposiums, and Workshops“) that you can easily access from the link on the left side of this blog.


happy engineers week!

Happy engineers week to all of my engineer readers!  (Actually, it’s almost the end of engineers week.  Sorry for the late notice–hopefully you can still find time to celebrate).  Today is Introduce a Girl to Engineering (and OR??) Day.  If you brought your daughter to work today, kudos!

The National Engineers Week Foundation describes Industrial Engineering as:

Industrial engineers determine the most effective ways to use people,machines, materials, information, and energy to make a product or to provide a service. Sometimes they are called “efficiency experts.”

Do you think of yourself as super organized? Do you think you’re good at understanding the big picture and figuring out how things could work better? If so, you might make a great industrial engineer. Your job would involve organizing people, places, equipment, and information, ensuring that complex and large-scale systems operate safely and efficiently.

For example, you might keep a hospital operating room running like clockwork. You might make sure an assembly line runs smoothly both for people and machines. Or you might be involved in adding a little extra fun and convenience to people’s lives by figuring out ways of making amusement park lines shorter, or by seeing to it that a big clothing chain always has every size of jeans in stock.

Although most industrial engineers work in manufacturing industries, they may also work in consulting services, healthcare, and communications.

Kinda sounds like operations research. Too bad they didn’t mention OR by name. Mike Trick wrote about the importance of the OR brand recently.  If you are an engineer and an operations researcher, how do you describe yourself?


don’t carpool!

Gas cost more than $4 a gallon until recently, which encouraged people to drive less, buy bicycles, sell their SUVs, and consider carpooling.  But is carpooling a good option? Sheldon Jacobson and graduate student Doug King tackle this question in a recently published paper in Transportation Research Part D (Transport and Environment), which made a bit of a splash.

They show that although carpooling saves gas, it guzzles time. For carpooling to become attractive, we would have to value our time at almost nothing ($4.24 for car drivers and $4.68/hour for light truck drivers). Car pooling makes sense if the carpoolers live close to one another (so they don’t waste a lot of time picking everyone up) and if they can save a lot on tolls and parking.

My husband carpools about three days a week, because he otherwise has a long commute (his work offers additional incentives for carpooling). The hardest thing about car pooling for him is the lack of flexibility in work times (he leaves early in the morning to return home at a decent time). If he isn’t driving, he can nap during the long drive to work. So if you’re a napper, car pooling can purchase two naps a day.

On a related note, a couple of years ago, Sheldon Jacobson and and I predicted that if lighter people from the early 1960′s drove today’s cars, we would save 938M gallons of gas a year due to the extra weight (both men and women gained an average of ~24 pounds between ~1960 and ~2002).  Sheldon Jacobson and grad student Doug King updated this research, since (1) Americans weigh even more now, (2) Americans drive even more now, and (3) Americans have grown taller.  Baesd on these factors, we could now save 1.21B gallons of gas per year if people from the early 1960′s drive our cars today (an increase of 21% over my estimate).

I found a couple of things to be fascinating when working on this research:

  1. Although men and women take the same amount of trips per day (4.1), men drive much more than women (37.6 miles/day compared to 21.2 miles/day). This is particularly accentuated for adults aged 25-54 years old, who do the vast majority of the driving. Say what you about how much we drive in this country, but most of the driving is done by adults going to and from work.  Unemployment obviously saves a lot of gas, but I can’t exactly see this as an upside to the rising unemployment rate.
  2. The extra weight in my car is due to car seats. Nearly every state has rigid car seat laws that require parents to lug around heavy car seats in their cars for years and years. Buying a lightweight car seat could save a lot in gas over its lifetime. One of my car seats is very heavy, and I feel guilty about this, but heavy seats often do well on the safety tests, and I am not willing to trade safety for gas.

What heavy items do you carry around in your cars that you can’t do without?  Do you carpool, and if so, why?


random guesses and intuition

If a student randomly guesses on a true false test, what is the probability that the student guesses at least 8 out of 10 correct? Before you get excited and start thinking of a Binomial(10, 0.5), listen to this SciAm 60 Second Science podcast. Researchers from Northwestern University published an article in Nature Neuroscience (key figure below), which indicates that “guesses” can be more accurate than educated responses.  Investigator Dr. Ken Paller, professor of psychology at Northwestern says:
“Remarkably, people were more accurate… when they had been distracted than when they had paid full attention. They also were more accurate when they claimed to be guessing than when they registered some familiarity for the image.”
Read more.  I am teaching stochastic processes this semester, and may have to work this research into an example. Of course, I might end up inviting trouble with, um, creative homework solutions, but how things really work can be fun in class.
The moral of the story: be intuitive.
Behavioral and ERP results from the Nature Neuroscience paper

Behavioral and ERP results from the Nature Neuroscience paper


depleting fossil fuels

My family is just now recovering after 2+ weeks of battling a nasty virus )-:  I missed blogging. I have something interesting to write about today.

Yesterday afternoon, I attended a talk by Dr. David Carlson (inventor of the amorphous silicon solar cell in 1974, currently at BP Solar) about the depletion of our fossil fuels, global warming, and solar energy. In the talk, Dr. Carlson guided us through the fossil fuel controversy, and encouraged us to think about photovoltaics and solar power.

Although I know little about peak oil, I’ll explain it how Dr. Carlson explained it in his talk, focusing on modeling. Oil discovery has been declining, and even the oil companies forecast declines. After oil is discovered, oil production ramps up, peaks, and then declines as the oil is depleted. Production peaks 25-45 years after discovery. Oil production peaked in the US in the early 1970′s and has been declining in North American (US + Canada + Mexico). Oil production has been leveling off in the Persian Gulf by some estimates. Has worldwide production already peaked? It may have. Based on (1) oil discovery and (2) oil production, many experts predict that the total worldwide oil production will peak between 2010-2020, but there is a lot of variation.

There are a few wildcards in these peak oil predictions:

  1. Unconventional oils could be widely used soon (decreasing demand for conventional oil), but it is possible that these will make even more carbon dioxide than conventional oil.
  2. All peak oil estimates depend on predictions on growth of the world economies.
  3. Cheap alternative energy sources could also make a splash, changing oil demand.

The bottom line is that predicting peak oil depends on several forecasting models (oil discovery, oil production as a function of discovery, oil demand, world economies growth). Each forecasting model has large error bars, and combining the sources of error doesn’t exactly make the predictions any easier (forecasting is tough!). But we are sure that (1) the planet is getting warmer and (2) we use a lot of oil, (3) our oil won’t last forever, and (4) carbon dioxide hangs around in the atmosphere for decades, so we need to do something now. It bothers me that the talking heads have been talking about peak oil forecasts as if they were certain, because the variance in forecasting estimates are generally very large.

Dr. Carlson also talked about how global warming is measured.  It isn’t easy.  All measurements have large error bars, and thermometers haven’t been around for very long.  In addition, although the average global temperature is increasing, it is not increasing everywhere at the same rate.  Arctic temperatures have risen much faster than the rest of the planet, resulting in fast melting of the the ice caps.

Measuring changes in sea level is also difficult.  Using tide gauges is problematic, since you cannot assume that land level is a constant frame of reference over long periods of time.  Satellites are currently used to measure temperatures and sea levels.

Listen to a podcast about peak oil. Learn about Princeton’s Stabilization Wedges.

I am sure that many of you know more about these issues than I do, and I apologize about any errors on my part.  Please leave a comment if you want to add something!


Follow

Get every new post delivered to your Inbox.

Join 2,423 other followers