Standardization is the enemy of Innovation

Standardization and innovation are ever in tension. Every rule we make hardens like clay, constraining the environment in which we operate, removing degrees of freedom. Every standard that gets baked in reduces the opportunities for innovation. This is a necessary stage as the ‘best’ rules get brought forward and the future is built on that.

Certainly standards lower the costs of deployment. Bespoke or custom designs for everything is expensive, while mass production of designs lowers costs. This is true not only of the resulting design, but the process of designing, the engineering itself. If every engineer had to rediscover physics and mathematics, construction would be costly. Capitalism and progress demand that we stand on the shoulders of those who came before, mostly ordinary sized people rather than giants, and reuse or adapt what they did, rather than reinvent from whole cloth.

The consequence of increased specialization along with standardization is that many civil engineers are in practice engineering technicians, following standard manuals and their electronic equivalents rather than creating new designs from first principles. On the one hand this is a necessary stage of mass production, bespoke design is expensive. We don’t have room for Isambard Kingdom Brunel anymore (and his lack of standardization was costly). On the other, it wrings creativity and innovation out of the system.

However the consequence is that more and more civil engineers (the engineers with whom I am familiar) are trained, in universities, to be technicians, to apply standards rather than develop unique designs. Learning how to properly use the manual or standard industrial software, undoubtedly useful and increasing employability in the short run, takes the place of understanding the problem from first principles, which improves practice in the long run. And when the manuals are less than scientific, or have fundamental problems  or have values gussied up as standards (ITE Trip Generation, Highway Capacity Manual’s Level of Service), this can make the world a worse place.

A feature of this process is that some (not all) civil engineering students seem to want to be technicians rather than designers. Many want to be told how something works rather than discovering it for themselves. They view their job as a calculator, not a creator. The expression “plug and chug”, (plug numbers into formulae and then chug out the calculations) in common use when I was a student, still seems to hold. And certainly, even students who are deeply interested in one subfield of Civil Engineering (say Transport) care little about another (say Hydrology). Though there are some common aspects between the two (flow on networks) there are also marked differences, and most engineers specialize, yet are required to take courses outside in subfields outside their specialization.

In Transport Engineering, the field with which I am most familiar, there are any number of  standardizing handbooks that take decision-making out of the hands of individual engineers, and into the handbooks (and thus committees of engineers who went before).

  • Transportation Research Board
  • AASHTO Green Book
    • A Policy on Geometric Design of Rural Highways, 1954 and 1965 editions;
    • A Policy on Arterial Highways in Urban Areas, 1957;
    • A Policy on Design of Urban Highways and Arterial Streets, 1973;
    • Geometric Design Standards for Highways Other Than Freeways, 1969;
    • A Policy on Geometric Design of Highways and Streets, 1984, 1990, and 1994;
    • A Policy on Design Standards—Interstate System, 1956, 1967, and 1991;
  • ITE
    • Trip Generation first (1976), second (1979), third (1983), fourth (1987),  fifth (1991), sixth (1997), seventh (2003), eighth (2008), ninth (2012), tenth (coming soon)
    • ITE Parking Generation first (1985), second (1987), third (2004), fourth (2010)
  • FHWA Manual of Uniform Traffic Control Devices (1935), (1942), (1948) (1961) (1971) (1978)  (1988), (2000), (2003), (2009)
  • NACTO Design Guides
    • NACTO Street Design Guide (2013)
    • NACTO Bikeway Design Guide second (2014)
    • NACTO Transit Street Design Guide (2016)
    • NACTO Global Street Design Guide (2016)

These are not taken as the word of the almighty by academics, there have been numerous critiques on Trip Generation in particular (See e.g. Shoup (2003)  JTLU (2015)), and the NACTO guides can be seen as implicit critique of the inappropriate use of the AASHTO books.

And certainly, we want some things, like the shape, size, and color of stop signs to be standardized so as we drive from one county to the next, we know where to stop (and so our robot cars are less likely to err as well).

I try to avoid teaching my students how to use a particular design guideline or handbook. They are generally simple enough students can get it on their own if they get a job where it matters. I used to tell myself it was important to teach first principles because particular design guidelines and software were transient, but they seem pretty ensconced by now, even with critiques as noted above.

I also know that not all faculty feel that way, and as a student many years ago, I had more than one course with units focusing on specific use of particular guides and employing specific software (at the time MS DOS based) to achieve particular results.

Will the revolutions in surface transportation (automation, electrification, sharing in particular) change this? Standards are lagging. Clearly the operational characteristics of AVs should smash the actual numbers in the HCM, HSM, Green Book, and so on. The methods may be slightly more robust. But I suspect we will be misdesigning roads for the last technology long after it has passed. Standards have consequences.