Artificial Intelligence for Engineering - How mature are we?

There's a lot of hype about Artificial Intelligence (#AI), Machine Learning (#ML) and what these technologies can do for/to the future of the human race, and continual research and development of intelligent algorithms for solving human problems is rapidly having an impact upon the world. In self-driving cars, recognising faces and emotions, synthesizing voices, and even making a hair salon appointment, AI has begun to make its mark in our personal lives. Yet, somewhat ironically, AI has had little impact on Architecture, Engineering and Construction today (I say this is 'ironic' because these fields are all based on STEM, logic, and rational thought).

Sure there's been a couple of cool things happening around the place, like automated progress monitoring robots, but these advancements are not all that pervasive, yet. So comfortable in the fact that their job is technically challenging, should the average engineering professional go home and rest at ease? Or is there an undercurrent that threatens to rise to the surface and in one-quick-move devour an entire industry? Disruption, after all, isn't a slow and steady process, rather it is a ruthless adversary that moves quickly throughout an industry and can wreak havoc in a very short time-frame.

If we look to the Banking industry, a very traditional industry now struggling with the emergence of very disruptive technologies, we can see how slow moving 'big business' is finding it difficult to adapt to a changing environment. So how will engineering firms react then when advanced AI's start eating their lunch? In fact I propose now is the time to look into it because truly;

• Gone are the days when we need 150 engineers, scientists and architects to design a 200 km stretch of road, or a 100 story high rise, when a small team of engineers and programmers can guide an AI to produce a code-compliant and fully-optimized design in accordance with the clients requirements.
• Gone are the days when we need a swathe of contractors to carry out all the civil works, steel and form work, fit-out etc. when we have an army of fit for purpose robots that can 3D print what we need, supply to and install on site.

These are but mere examples of a new form of consolidation, and unlike mergers and acquisitions which aggregate body counts and the skills and capabilities of human capital, this consolidation is happening in the cloud, and it's all about enabling thousands of pop-up AI-driven service providers to start taking small, but meaningful bites out of our sandwich.

"Yeah right" the engineers say in disbelief, comfortable in knowing that the past 30 years of their professional career has instilled in them knowledge that no AI could possibly compete with. I mean, thousands of years of scientific advancement has culminated in their ability to achieve great things, never thought of in the past! Just like thousands of years of accumulated knowledge of the art of playing Go, a game which has more potential moves than there are stars in the known universe meant that even the most complex AI couldn't... Oh wait! that's right, after only a few hours of training, an AI beat the entire knowledge base of human existence and a lifetime of experience and training and turned the game upside down.

Where are we now?

Lets entertain for a moment, the idea that an AI will one day be doing our jobs. Its not hard to believe; much of engineering is about applying well understood rules to the generation of a solution. Take a recent project I worked on in which we developed software that carried out an entire design which traditionally took 2 weeks. The software, programmed to understand the rules, completed the task, error free, in 30 seconds. There's value in efficiency and in getting the right answer the first time, and even greater value in getting a correct and optimized design right off the bat. Now I hope you're thinking, "ok, I'm interested, so what?"

Well.. great! The first step towards solving any problem is realising that we have one, and where we should be applying our efforts towards a solution. So, lets start with an assessment of where we are right now.

For the past month I've been developing a framework for the engineering industry against which we might assess our current AI maturity. Its pretty straight forward, we simply consider the most recent projects we've been involved in, and the extent to which these statements represent how we delivered those services.

Where are we going?

So we exist somewhere along a curve, but knowing where we are is only one part of the equation, lets also look at where we're going. We need to think about our business and imagine for a moment we're 10 years in the future. Robots are pretty much doing everything, artificially intelligent personal assistants are connecting us with new services, our business is going great because we've fully embraced AI and are leading the industry. So what are we using AI for? What things that humans are doing today, is the AI now doing for us? Has it automated part of a job, or everything we do? We are in control, and we have skilled people working around us; what jobs do they have in this new world?

There's a lot to think about, but the clearer the vision we have, the easier it will be to discover the next steps to getting there. What's more, there's a great way to do this, I've been working with clients for the past year on creating a vision for their future, and a few simple facilitated workshops with key stakeholders in our industry and business is all we need to get started.

How will we get there?

Now that we have a vision, there's a lot of work to do, and we need to put in place a plan of action, it takes into account not only the gaps in our current capabilities, but also the changes to our business model. Moving to the world of ubiquitous AI isn't about dropping everything we do either. Some of the same systems, processes and people we have today will be critical to our success, and there are ways of taking people from the organisation along on the journey with us. While there is no one-size-fits-all solution to getting from today to tomorrow, with the right advice, a positive attitude, and supported by a clear vision from above we'll be on our way in no time!

Final words

First and foremost, thanks for making it this far, I hope the framework is useful for discussion, and I'm open to comment and feedback on it. Where do you sit on the scale? As a practitioner of parametric and generative design and with experience in the application of ML algorithms to design optimization, I'm always looking for ideas on how we might advance the field further, your participation in a discussion below is a great way to achieve just that.

Secondly, as a management consultant, I know the difficulties many people and companies face when considering the changes they need to make to adapt to this rapidly changing world. I would love to have a one-on-one conversation with you about how you can position your company, and yourself, in this new world. I'm happy to present to you about the opportunities that AI might also offer across a wide range of industries.

Finally, please feel free to use the framework for your own purposes, I only ask that if you do make some changes, please give me feedback so I can improve it for the greater good.