Saturday, September 24, 2016

Self-Driving Cars and Nuclear Power, Part 2:

Echoes of Familiar Themes

When I started my previous blog, I realized that I had also stumbled upon another theme about self-driving cars that echoed the history of other technologies, and that is, predictions about the impacts of the technologies.  In this case, the predictions were about whether self-driving cars would help the environment or harm it.  The answer seemed to be a resounding YES!

The article really highlights the dilemmas of trying to make predictions when the final outcome depends not only on the technology, but on the rules that will govern it, and on how individual and collective behaviors will change.

In this case, self-driving cars conceptually will not need all the safety equipment we have on current automobiles.  This should save weight, and therefore, save fuel.  What the article doesn't address is whether the public and the regulators will even feel confident enough to remove requirements for safety equipment.  We have already seen one fatal accident in a self-driving car.  Will people want the anti-skid brakes and the airbags "just in case"?

But, say that we do eliminate some of the requirements for safety equipment, resulting in a significant energy savings.  The other side of the coin is to wonder what impact a self-driving car will have on individual behavior.  Will people move farther outside of town because the commuting time will no longer be wasted time, thus negating some of the energy savings.  Or, bizarrely, will people let their cars drive around endlessly in cities rather than paying for a parking space!  Or if they can't find a parking space.  (I shudder at the road congestion that would cause!)

The article raises other issues as well--will cars be permitted to drive faster, which is less fuel-efficient; will they be programmed to follow the most fuel-efficient route; etc.

Again, this dilemma reminds me of the fact that automobiles were initially perceived as being a great boon to the environment.  After all, they would eliminate all the pollution from horses on city streets!

Once again, at the very beginning of a technology, it is hard to envision all the ramifications a technology should have.  That's no excuse, of course.  This article does a service in pointing to some of the issues that need to be addressed in parallel with the technology development and the testing.  Further thought needs to be given to other potential impacts self-driving cars could have.  So far, most of the emphasis has been on the greater safety, and that is truly a big factor, but more thought needs to be given to the unintended consequences on energy use, road congestion, urban and rural communities, etc.

The same, of course, can be said of any new technologies.  In fact, another prime example would be power plants for electricity production.  A small number of any type of power plant has very little impact, but once you ramp up the percent penetration, you need to think more seriously about the amount of land required, the materials and manufacturing processes needed and their environmental impacts, the implications for the transmission grid and backup systems, and the impacts on air and water quality.

No technology is going to be perfect.  No technology is going to be optimal in all respects.  However, anticipating and planning for new technologies can help society make informed determinations about which technologies to use, where and how to deploy them, and how to minimize their negative impacts.   


Saturday, September 17, 2016

Self-Driving cars and Nuclear Power, Part 1:

Echoes of Familiar Themes

No, I'm not going to advocate nuclear power for self-driving cars!  I was just struck by a couple of articles I've seen over the past few days that demonstrated to me some fundamental similarities among diverse technological developments.

In the first case, someone pointed me to a couple of articles in IEEE Spectrum about the history of self-driving cars.  One article cites reports dating as far back as 1958, while the other references a previous IEEE Spectrum article in 1969.  Both articles are optimistic about the future of self-driving cars, and to read them now, one would have thought the technology was around the corner.  And this was before all the modern computer and other technology that is being applied to the current demonstration models of self-driving cars.

As it happens, these two articles were sent to me just days after someone else sent me a list of predictions being made today in a variety of areas.

My point here is not to delve into the the pluses and minuses of all the recent predictions, or to try to assess the details of self-driving cars, then and now.  I am not an expert in most of these areas, and will leave that assessment to others.  But what did strike me is that elements of both these discussions seem eerily similar to the expectations for nuclear power in the early days.

When I worked at DOE in the early 2000s, I often gave talks on the Generation IV activities we were then just beginning.  One of my points addressed the "new" nuclear reactor technologies we were planning to pursue, such as molten-salt technology.  I remember one memorable evening when the first person that I recognized in the Q&A session after I completed my talk chided me gently for calling molten-salt reactor technology "new."  He himself, it turned out, had participated in the early work on molten-salt reactors--some 50 years before my talk!

And furthermore, that work on molten-salt reactors was associated with an ambitious program to develop a nuclear-powered aircraft, an effort that, for a number of reasons, never came to fruition.

All of which brings to mind that famous quote:  "It is hard to make predictions, especially about the future."  (I had always seen this attributed to Yogi Berra, but I have recently seen claims attributing it to a number of other people, including Neils Bohr.)

This is not to say that all predictions are wrong.  Far from it.  Clearly, the successful implementation of a ground-breaking new technology depends on a number of factors.  Sometimes, the technology just isn't advanced enough.  Sometimes it costs too much.  Sometimes there is a better alternative introduced at the same time.  Sometimes the need isn't there yet.  Sometimes, politics or public opinion or unexpected current events create barriers.  Sometimes, some or all of these factors change over the course of 50 years, and what didn't succeed before ends up succeeding the second time around.  Or the third time.

I personally am very hopeful about this current generation of self-driving cars, and I am very hopeful about the current round of development on advanced reactor technologies.  But I take heed of the lessons of history to note that all new technologies face challenges and potential roadblocks, some from the technologies themselves and some from other sources.  This is not pessimism, it is realism, and it suggests that proponents and advocates need to understand all the complex factors that may affect their plans and try, wherever possible, to address them in a timely fashion.  

I can only end by invoking another quote, this one, I'm pretty sure is only attributed to Yogi Berra:  "It ain't over 'til it's over." 


Tuesday, September 6, 2016

Women and Engineering:

Still a Long Way to Go

I started my education, and a few years later, my career, in an age where women were still a relative rarity in the sciences and education.  My undergraduate class at MIT was only 5% women.  As a graduate student, I was the only female student in the entire nuclear engineering department (at that time, only a graduate department).

Therefore, I tend to view the current statistics on women in science and engineering as nothing short of miraculous:  Recent MIT entering classes have edged near 50% women.  Wow!  I am now seldom the only professional woman at a meeting.  Super!

Thus, I read with interest an article in the Harvard Business Review noting that many women who enter the engineering profession end up leaving the field.  As a result, while considerable progress towards equal representation has been made in other technical disciplines, nearly 40% of women who study engineering never enter the engineering workforce, and women still constitute only 13% of the engineering workforce.  By contrast, the article points out that the number of women in medicine and law is approaching the number of men in the field, and the number of women in the basic sciences is on the rise (although they give no numbers).  Clearly, I had been looking at one set of numbers that looked very promising, but digging deeper, the picture gets murkier.

The article cited several factors behind these trends.  They focus most on subtle forms of discrimination in team-related and internship activities.  These include things like the teams assigning the female members the more routine, non-technical tasks.  Certainly, being cut out of the most interesting work must make engineering seem like a less exciting profession.

The article also indicates that women, more than men, seek "socially responsible" work, and they feel they do not find that in the engineering profession.  I must admit that I find this argument a little puzzling, as I've always felt that engineering professions offer more opportunities than most professions to provide direct benefits to society.  While it is not the only profession that benefits society, it offers a broad range of possible ways of improving people's lives and welfare, whether by helping provide more energy, cleaner water, safer transportation, better labor-saving devices, or in a host of other ways.

I know that some will observe that the study doesn't say much about what happened to the women who left the engineering profession.  Did they, indeed, end up doing work that benefited society more directly than they could in most engineering positions?

While I can agree that would be nice to know, the fact remains that in the year 2016, we still have behavior in academic settings that effectively sends a message to women that they are not equal, and that takes away from them the full opportunity to contribute to projects and to enjoy the challenges of the profession.  This doesn't do anyone any good, and I hope this study causes the academic community to look more closely at the team projects and how they are run, and to seek ways to address the types of behavior the study found.  In addition, I would hope that the professors and others make an effort to explain the value of engineering activities to society.  It is apparently not as obvious as some of us in the profession think it is.