Getting $$ from Process Data
Dr. J. Patrick Kennedy, CEO of OSIsoft, provided the closing keynote for WBF 2008.
PC Vs IT
-- IT Does it Right
-- IT Doesn’t Matter
IT
-- Part of All Things Modern but No Stand Alone Value
-- Peer to Peer Redesign
PC
-- Difficult, Treacherous
-- Stand Alone No Value
About OSIsoft
-- Based on Idea Plants needed Real Time Data Mgmt-- something people said wasn't needed...
-- 28 Yrs Old
-- 2008 - $250MM/600 People
-- PI Killers
Secret
---- Listened
---- Focused on Value
The Math of Success
--- $$ IN
--- $$$$ OUT
--- High Expected Present Value
The people upstairs are looking for projects that will provide high current value at low risk. There are enough troubles running a public company...
Reality
--- Mythical Man Month
--- Low Hanging Fruit
--- System – Continuous Improvement
--- Improvement over Time
Three Examples:
Batch Digesters
Steel Mill
Chemical Park
The batch digesters...1% improvement is worth 1% of the value of the mill! You have to be aware of the value of what is being done...NOT the COST!
The steel mill...steel mills are industrial archaeology havens...it is sometimes hard to tell the scrap metal from the process units. They break a lot...and they were the typical maintenance people...fix it when it breaks. This doesn't give you time for improvements or engineering. Proactive maintenance...you do the things you have to do to make things work. They went from 70% reactive maintenance to 20%. Was that worth anything? 16 % improvement in availability of the steel mill. Yes, these things are worth the money.
The chemical park...Kodak. And you all think you are in a bad business... try making film.
1300 Acres
>20,000,000 Square Feet
11,000 Employees
30 Miles Road
Fire Department, Railroad
Water and Waste Water Treatment Plants
Operates Two Power Plants
They went from three distilled water plants to one, firewater went down, reducing catalyst regen costs, and all sorts of stuff like that. They moved their power peaks to where they were paying negative dollars.
Kodak got savings in every department:
Water Management
Environmental Regulation
Asset Management
Troubleshooting
Operating Effectiveness
Energy Management
Inlet Vanes
Utilities (Power, Steam, Air)
Trading/Power Purchase
Lessons Learned
No BIG BANG – 1000 little bangs
Continuous Improvement Process
Combined Capital and Intelligence Operation
Infrastructure Approach – Remove Infrastructure from Projects
Value Monotonic w/Scope
Lowers the Cost of Curiosity
To Get the $ You Have to Deliver $$$$
In March of last year, Kodak shut down an entire power plant. That system is now making $28 million a year for Kodak.
World Competition
No long cheap raw materials/labor
Biggest market in the world
Compete on Strategy and IP
RT Data Management New
Great Way to be Seen
What has to happen is you have to create a forum (perhaps at WBF) where you can share IN DETAIL that you can PROVE where you can make money for your companies. Where users can come and talk to each other about where the benefits are. Your management will come up with reason after reason why you should not do these projects...so you need rigorous papers that allow you to prove your case.
Forget the technology...they don't care if the models are linear or not...they don't care about S88...they care that what you do works and makes the company money...that's your job. What they need from you is how they're going to pay for it. Scare tactics don't work. Security doesn't sell. Safety doesn't sell. What sells is proving that you can make your management more money.
Performance based pricing never works, because by the time the company sees what the value actually is, they don't want to pay the price.
Piloting really doesn't work either...do it first.
