MIT tool aids cost estimates for complex projects
On September 14 in a blog called The Pipeline I speculated about how to develop a number of the pieces for a 1000 mile long pipeline. I had a lot of fun anyway. But really, how much would such a project cost? Last week I discussed Changing World Technology’s thermal depolymerization process — which today produces diesel fuel profitably from biomass–including city sewage– with fresh clean water as a byproduct. But once again, it would be nice to have a breakout on its costs for any given particular location and feedstock. There are number of power plants up and down the California coast for which a number of different technologies can be used to harvest waste heat and waste carbon dioxide to make fresh water and energy. It would be nice to have a tool that would enable you to easily cost out the variables. All over the west oil men are pulling up brackish water with their oil. How much would it cost to clean up that water for public use per any given location. Once again it would be nice to have a tool that answered that question. Then there’s the big research question: How much would it cost to develop a cheap low maintenance semi permiable membrane that desalinised water at room temperaterature and pressure.
Likely a company like GE has some kind of in house software that they use to predict costs of big projects. But what about everyone else? How do people comfortably bid on big projects. How do government agencies evaluate those bids. MIT News put the problem this way in this article entitled
MIT tool aids cost estimates for complex projects
Michelle Gaseau, Lean Aerospace Initiative
September 19, 2006
Consider the following scenario: A project manager at a major aerospace company is about to bid on the development of a new air fighter for the U.S. Air Force.
The bid must bring the project in on time, on budget and meet all the government’s requirements. If the bid is too low, the project will miss these markers; too high and the company will be seen as wasteful or inefficient and may disqualify itself from the competition.
Now a new, first-of-its-kind systems engineering cost-estimation model developed by an MIT researcher can ensure that the bid is right on target, which means project risk (and costs) can be reduced. The model allows companies and organizations to develop more accurate bid proposals, thereby eliminating excess “cost overrun” padding that is often built into these proposals.
The software takes the guesswork out of bidding on projects and allows government administrators to effectively evaluate those bids.
The Constructive Systems Engineering Cost Model (COSYSMO), now available commercially, helps eliminate the guessing game played by many large corporations in planning and executing large systems in many different industries. It also helps government agencies evaluate proposals from contractors with a more objective approach.
“In the past, a program manager would look at an earlier aircraft program and estimate by analogy, but now we can go beyond that and use parametrics to go beneath the surface to the underlying reasons why a certain aircraft costs what it does to develop,” said Ricardo Valerdi, a researcher at MIT’s Lean Aerospace Initiative (LAI) who developed the new model.
This software is good for not only aerospace but also any large diversified project.
Validated with assistance and historical data from seven major aerospace companies, COSYSMO can be adapted to systems engineering programs in many different industries.
“The inputs to the COSYSMO model are generic, they are not domain specific, so it could be used in estimating effort associated with waste management systems or building new highway tunnels in Boston,” said Valerdi.
That means pipelines, — desalination and dual purpose plants would also be appropriate for this software.
Systems engineering is an interdisciplinary approach to creating successful systems by focusing on variables including customer needs, system requirements, design synthesis and system validation all while considering the complete problem.
According to the article — this is a very new thing. On this I have my doubts. But cost over runs do seem to be a way of life in the government. And this software does claim to solve this problem.
Others have developed cost-estimation models for computer hardware and software development, but until now no models have been created to estimate the costs associated with systems engineering.
Computer hardware and software cost-estimation tools help companies estimate costs specifically associated with developing and designing computer hardware and software components and platforms. The costs associated with systems engineering are more difficult to estimate because the discipline deals with multiple factors in the big picture such as system design and customer needs.
COSYSMO helps companies estimate “person-months” specifically associated with a systems engineering effort and costs — such as how many people it will take to develop a command and control system in an aircraft and meet all the customer requirements.
According to Valerdi, the failure to adequately plan and fund systems engineering efforts appears to have contributed to a number of cost overruns and schedule slips, especially in the development of complex aerospace systems.
In fact, some of the companies most famous for cost over runs participated in the validation of the new software.
In addition to its availability via commercial channels, the academic version of COSYSMO and its new user’s manual are both available to members of the LAI Consortium. Many of the consortium members, including BAE Systems, Northrop Grumman, Lockheed Martin, Raytheon and L-3 Communications, participated in the validation of COSYSMO.
Three corporations now offer COSYSMO commercially: Price Systems, Galorath and Softstar Systems.(my note: Softstar has a free trial vers
A version of this article appeared in MIT Tech Talk on September 20, 2006 (download PDF).
It would probably be a very good thing if desal people in each of the DOE, the EPA, the DOD, a couple Federal labs, a California, Texas & Florida water agency, a couple desalination non profits like WaterReUse, American Water Works Association and the Water Environment Federation, an oil driller–as well as GE and a couple smaller desalination companies…picked up a copy of this software, each hired a $90k year staffer to follow around some old water desalination/energy systems engineer goober and adopted his knowledge to the software.
That would be an ugly sight. So it might be best to hire someone for 60k who will go around and teach everyone to use the software. In any case with enough diverse people working off the same software cross checking each other and building a library of cost-estimation models — the industry would be well positioned to react fluidly to rapid changes in technology and, indeed, the world.