There have been some troubling signs: a 2010 article in OR/MS Today suggested that analytics would subsume operations research; a 2013 LinkedIn discussion asked “Will Big Data end Operations Research?”; and ominously, even INFORMS seems to be distancing itself from operations research.
How should we react to this? Should we:
- Take early retirement, move to Vermont, and open a bed and breakfast?
- Claim to be analytics professionals, and hope no one asks us about Hadoop or NoSQL?
- Return to school to study data science?
No! None of the above will be necessary. To understand why, it is necessary to go back to first principles.
The original meaning of the name operational research (what operations research is called in Great Britain, where it was invented) was literally, scientific research on operations. The name was meant to distinguish scientific research on operations, from scientific research on the underlying technology of some product, e.g. radar. In the late 1930‘s the British Government funded scientific research directed towards creating radar equipment with sufficient range and precision to locate attacking aircraft. They also initiated an operations research study to determine the most effective way to deploy the radar stations, and integrate them into an effective air defense system.
This type of scientific research, and the scientific method upon which it is based, is a problem solving paradigm. Operations research is the application of this problem solving paradigm to the solution of operational and management problems.
During the summer of 1940, this paradigm arguably saved Great Britain from defeat. Today, as the Edelman Competition routinely demonstrates, this paradigm creates benefits so great, that they transform entire organizations. And, it is because of this paradigm that operations research can create value that can be created in no other way. This value — lower costs, higher profits, military advantage, more efficiency, better service — was needed in 1940, is in evidence all around us today, and will be in demand for as long as human civilization persists.
So, there is no cause for alarm. Just continue ‘Doing Good with Good OR’.
It doesn’t matter how good your Big Data and analytics are if they don’t lead to better decisions. There seems to be a built-in assumption that more and better data will lead to better decisions. In practice, people are poor at making coordinated sets of complex decisions, and often adding more data actually impairs their decision-making.
OR is about the science of making better decisions. Someday the rest of the world will begin to understand this. It might take a year or a century; but without better decision-making, all the big data and analytics is just noise. I’m not giving up – big data and analytics are bringing new opportunities for us all. These are tools that we can use. After all, how often have we been hampered by the lack of good data for our models…
Robert … Great observations. I enjoy your blogs.
I view Big Data as a huge smorgasbord of ingredients and OR as selecting the right ones to prepare a healthy and delicious meal.
Gary … Gary Cokins
Hi there
Thanks for addressing this issue. I’d like to believe you, because I’m starting my career in OR now. But I think we underestimate the power of Big Data to remove the necessity to make good human decisions. Lots of decisions that previously would have needed OR techniques now do not need them because of the speed and accuracy of programs designed by developers/Big Data people. For example, you don’t need to schedule maintenance for your assets if you can get real-time information about their health and be able to crunch it in a fast and informative way.
I think the role of Data Scientist appears to me to be an experienced programmer who knows most of the OR techniques, *and* knows their Hadoop from their NoSQL. And where one could claim that an OR-trained person is important because of the wide selection of techniques they know, I think this is becoming redundant because the skills of economists, research scientists and top business people are becoming good enough so that they can get 80% of the way there without needing to consult an OR-trained person.
Yes, OR is about the science of making better decisions, but I think the new breed of Data Scientist knows how to make better decisions too. I think that Data Scientists have the skills to set up smart programs that can handle Big Data in a way that an OR person cannot, and so open up new avenues for decision-making.
There may still be a role for OR in the distant future, but I think it will be reduced, and I think skilled developers and data scientists will provide much more value for companies looking to make smart decisions. In the mean time, I’m going to try to learn some more techniques, and find companies which are immature on the data science spectrum, and try to make a difference there.
I think it is easy to overestimate our usefulness as a skillbase. I think looking backwards in history to explain why OR will be useful in future is naive – the world is a very, *very* different place. I think it would be better to look to the future to decide whether OR will be useful in future. Does OR help more than data science when the world is about super fast information, app-based solutions, and Google? I don’t think OR has anything special that other fields don’t have already – I am not comfortable claiming that OR adds value in ways that nothing else does. It might be so in a minority of cases (I can’t think of any – please give me an example?). I think OR matches up poorly against things like data science in the areas of crossover between OR and data science. And I think any areas where we can claim total supremacy of other fields will diminish in future as the power of Big Data matures.
If you can change my mind, I’d be very happy to hear it. I would love to hear about OR techniques and problem areas where you feel that OR will continue to provide more value than any other field.
Cheers
Paul Edkins
Hi Paul,
I urge you to look at the abstracts of high impact operations research projects that can be found by clicking on the following link. (Abstracts are present for all projects up to 2011.)
https://www.informs.org/Recognize-Excellence/Franz-Edelman-Award
I also recommend that you review the papers in the INFORMS practice journal ‘Interfaces’. You will find complete papers describing all award winning projects, as well as many other successful operations research projects. These papers will give you an idea of the scope and power of operations research.
As for areas where operations research has had a major impact, I can offer the following partial list: airline route planning; scheduling airline flight crews; revenue management; supply chain optimization; inventory management; forward buying optimization; scheduling part-time employees; designing telecommunications networks; military logistics; combat modeling; location analysis; production planning and scheduling; vehicle and fleet routing; allocation of resources; planning effective and cost efficient customer service systems.
Finally, the rapid changes in the business environment lead to more opportunities for operations research to be used to solve difficult problems, and the rapid advance of computer technology and the increase in data availability make it easier for operations research to solve those problems.
I think your first paragraph almost perfectly illustrates the big data fallacy. Having real-time and maybe even perfect information about the status and health of your assets absolutely DOES NOT tell you when is the best time to maintain/repair/replace your assets, unless you are in a few specific cases (e.g. only a small number of long-lived assets, no SLAs, etc). It sure would be nice to have that information, and it’ll help me (or my clients) a lot, but *all* it tells you is the health status of your assets. That still leaves open the questions of how to decide when to service those assets, whether to repair or replace, who does the job, where do we get the necessary parts, etc.
As for examples of where OR has helped, think about almost every service you use. Water, gas, electricity, sewage, postal and delivery, roads, warehouses, supply chains for Amazon etc, Some of it has grown up in an ad hoc manner, but usually there has been somebody with an OR background behind the scenes helping make those long-term planning decisions and setting up policies, inventory replenishment strategies, capacity planning etc. When you take a flight, do you think about the aircraft much? Where has it come from and where will it go after you get off? Planning the sequence of flights for a fleet of aircraft to cover the flight schedule is a complex process as it has to also fit with air crew schedules, preferred places to buy fuel and many other constraints and preferences. Absolutely sure that BIG DATA will help us. But even having perfect real-time data on everything does NOT make your decisions for you.
Do not confuse DATA sciences with DECISION sciences. One is about how to gather, process and refine DATA to extract meaning, and the other is about how to make DECISIONS. These are not the same.