FEED & Basic engineering 1

[greg211]

hi all,
i would like to thank everyone who responded to the earlier question i posted about spec break.Now to the issue at hand.
what is the difference between FEED & Basic engineering. which of them comes first before the detail engineering phase.

 

[BigInch]

This is a pretty good description that should serve to get you front end oriented,

http://www.innotec.com/feed.html?&L=1

http://virtualpipeline.spaces.msn.com

 

[JLSeagull]

See the contract. The operating company prepares the scope of work for a FEED or Basic Engineering project. That scope document defines the requirements for that phase of the project. Some basic engineering projects include the instrumentation data sheets and DCS / SIS requirements that are suitable to issue to vendors for a quotation. Some are more conceptual. It depends upon the scope defined in the contract.

 

[MortenA]

There is probably many opinions on this subject.

Biginch I think the description in your link could be one - but on the other hand i find that +/- 25% cost estimate without P&ID's is overly optimistic.

Its easy to claim this - but how are you going to document it? On the other hand you will almost guaranteed not be kept accountable to it - since so much will change that your estmate wont be comparabel to the final product. But i'm not sure thats good engineering. But maybe +/-35 would be more realistic.

Best regards

Morten

 

[BigInch]

I wouldn't say its hard to do at all, if you've done a couple of similar jobs during the last year and you are going to get budget quotes from your usual contractors. You did put a plus and minus in front of that 25, so that's really 67% over the minimum bid. Its been a very rare event when I've ever seen bids outside that range (hi >= 1.66 x lo) on a major project, but I suppose it depends on your company's or engineer's recent level of experience with similar work and how well developed your design is at the time of releasing for bids. If you go with a low experience contractor on a quick bid project, you'll probably have to increase your budget to compensate for the, shall we say, corresponding "unknowns".

http://virtualpipeline.spaces.msn.com

 

[GregLamberson]

greg211

FEED & Basic Engineering are simply descriptions of the same basic function, different companies call them by different names. Generally in a project timeline you will have the Assessment phase during which you determine project feasibility and alignment with business strategy.

Next comes the Evaluation Phase which is basically broken down into 2 basic sub-pohases: 1) select the concept and complete preliminary application filing requirements (i.e. FERC and NEB), and then 2) submit and support permit application and progress the project to the level required for the Definition Phase (Detailed Engineering). This is what is generally referred to as FEED or Basic Engineering.

After you go to Detailed Engineering, Execution, and Operations.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[ifreeman]

Sometimes they mix and use those terms in case of single unit project such as simple separation column.
And in such a case, there is no difference between them virtually.

But in case of mega project consisting of various process and UT/OS units, it's totally different story.
Usually the FEED contractor shall perform basic engineering work for several units if required.
And collect information from other licensor and/or basic design company as well.
And they make a FEED package so that EPC contractor can bid based on the package.

So, FEED package shall be accurate and well-defined so that the bidders(EPC contractors) can estimate their cost.

But as I mentioned above, for a small project, Basic Design Package can be used for EPC bidding.
So there is no difference between them.

ifreeman.

 

[MortenA]

Biginch

Cost est. isnt my strongest side (usually only provide input) and im sure im following you:

Quote from you:
"you are going to get budget quotes from your usual contractors. You did put a plus and minus in front of that 25, so that's really 67% over the minimum bid. Its been a very rare event when I've ever seen bids outside that range (hi >= 1.66 x lo) on a major project"

But you _dont_ have initial quotes. I mean isnt that excactly what the FEED is supposed to get you?

Im not talking minor modification of existing a plant or e.g. duplication of existing units or so - but grass root projects 100-200 mill euro.

Best regards

Morten

 

[BigInch]

Right, my only point was, if you did a similar project during the last 2 to 5 years, you'll have enough experience to guess a cost number within 50% of its final value... I hope. The +/- 25% really means a 50% range of the project cost. That's not a difficult target to hit. While projects do vary in some specific details, the law of averages rule in the end. Its surprizing how accurately you can estimate a "standard" (major) project just using say 1MM/pipeline-mile or 1MM/BBLD of refinery capacity and some $2K/installed horsepower for a large pump station, $5000/kWp solar power station capacity, or even $1MM/ton-year paper processing capacity. Throw in a little experience factor for work you or your company has already done in a similar area, the USA, Mideast, North Sea, Far East, factors etc. and you'll easily nail it somewhere within the actual high and low bids that you'll get later. Just check the OGJ, Pipeline Industry or similar industry specific publications for the current year's base cost factors. Use your head and your experience and you'll not have a problem hitting the mark.

http://virtualpipeline.spaces.msn.com

 

[GregLamberson]

Exactly BigInch, when I see a +/-50% estimate request I have to just laugh. It's ridiculous. +/- 25% is more in the ballpark, anyone with enough knowledge of the industry to be asked to put together an estimate should be able to get pretty darned close.

Early cost estimates will range in both accuracy and, depending on the Owner Company, by description. They are often called “screening”, “order of magnitude”, “+/- 40%”, “conceptual” etc.

To improve the understanding and expectations of the cost estimate, the Owner should adopt a standard classification system that relates the available technical data to the level of cost estimate accuracy. As a project becomes better defined, the estimate will become more accurate and contingencies/allowances used will become more targeted.

It should really be avoided calling a cost estimate as “+/- 40%” as this is rarely the case and often causes disagreements. If a level of accuracy must be put on it by using a “%”, it must be fully understood what is meant. Better to use terms such as “order of magnitude”, “rough estimate” or a similar description. More importantly is to understand exactly what the Owner wants to accomplish.

The key point is that the cost estimate must be tied to an internal understanding of the level of project scope definition that has been developed; the objectives of the study are known; and, that the expectations are fully understood within the Owner’s team. This allows the estimator to understand the context of risk, uncertainty and inaccuracy in a conceptual estimate so that the full range of potential outcomes of the estimate can be used in the most intelligent and useful way to aid in project decision making.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[MortenA]

Greg

If its that easy how come that so many projects ends up costing a LOT more than +40% of intial cost est (not some !revised cost est later on"? Is it because the clients consistently ask people who dont know?

Best regards

Morten

 

[MortenA]

Greg please dont get this the wrong way i would really like to know.

I agree that (quote) " key point is that the cost estimate must be tied to an internal understanding of the level of project scope definition that has been developed; the objectives of the study are known; and, that the expectations are fully understood within the Owner’s team. This allows the estimator to understand the context of risk, uncertainty and inaccuracy in a conceptual estimate so that the full range of potential outcomes of the estimate can be used in the most intelligent and useful way to aid in project decision making. "

But do you feel that PFD's+gut feeling (== experience) is sufficient for +/- 25%?

Best regards

Morten

 

[GregLamberson]

MortenA

Ah, now you are getting to the crux of the matter. But let's separate a few things here - the differences in the estimate and final cost. There are many factors involved that drive the final costs of a project that cannot and should not be taken into consideration but the estimator.

There are contracting strategies that have a huge influence on the project cost & schedule. Market conditions can change. Client organizational skills (or lack thereof). Contractor/labor availability. Internal politics play a part. I consulted for an oil major on a huge project in Russia, the PMT decided to exclude contractor profit and contingecny in their cost presentation to sr management becasue they wanted the project to move forward and thought it might get stalled (I kid you not). Stakeholders can influence cost (like in a PSA). Governments can change contractual terms.

None of the above should be taken into consideration by the estimator unelss specifically told to do so. The estimate the estimator puts together should show a middle of the road approach along with the estimate basis. The estimate is not the issue, it's how the project is managed in all regards that influence the final cost. Getting an estimate close is not a problem, the way a project is managed blurs the line between estimated costs and final costs. But it should not matter, a management team that receives an estimate must take that estimate and risk it, adjust it with and for a variety of factors. That's all out of the hands of an estimator.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[GregLamberson]

MortenA

Sorry, we posted at the same time - yes I do think that with PFD's, experience, AND a good cost model will get you within a +/- 25% cost estimate or at least close enough for management to be able to make a decision to continue pursuing the effort or not.

It's amazing how close early estimates can and are at times. The problem is one of risk - management, financers, etc. do not want to take the full risk based on an estimate tied to a PFD. So they will spend a few million on a Feasibility Study to get their confidence level up. But as BigInch pointed out, these are early estimates are estimated based on data bases of actual intsalled projects, so yes, they are pretty darn close.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[MortenA]

Thats an etremely good point

Some of my colleagues more into cost estimates than me in a resnt project used something they call "Monte Carlo" simulation. For input data on equipment etc. we had to supply our best guess and then our best guess for maximum and minimum cost for various major equipment, labour costs, etc.

This was then used to produce an estimate where the base price was the 50% fractile and the estimate would then proceede to state the 95% fractile (it will only be more than X% mere expensive in with 5% likelyhood) and 5% fractile (but it will not be less than Y% with 5% likelyhood). As far as i heard the client had some difficulties accepting this. While the 50% number becomes a litlle meaningless - i think that the upper/lower limit on the other hand makes sence.

I liked this approach a lot because it (at least) incorporates the known uncertaincies. If you have estimation for some of the other than you mentions then this could be factored into the model too. I allways have a little problem accepting these standard cost estimates with a number - and then with not much further dokumentation its +/- ZZ%

Best regards

Morten

(long post hope it makes sence when i press send)

 

[MortenA]

To your latest post: OK i tend to agree with you when you walking in familiar terriotories.

Morten

 

[GregLamberson]

Exactly, there are several methods for determining the amount of contingency or risk needed to be added, each with advantages and disadvantages. These include estimator judgment, percentages based upon the level of engineering detail and class of estimate, Monte Carlo analysis, or a parametric statistical approach. Probabilistic analysis is used to calculate expected values to understand the potential range of outcomes and form a decision policy. The tools to do this are decision trees, cumulative probability distributions, and Monte Carlo simulations.

Monte Carlos simulations are a methodology. What is commonly done is to run the cost model applying the range for all the risks assessed and accounting for the identified interdependencies. This, as your colleagues mentioned, can be done with a Monte Carlo tool such as Crystal Ball, DA software such as SupertreeR, or DPLR (Decision Programming Language) in which the various P10-50-90 values can be sent from the software to the cost model and the resulting total cost calculated.

That's when sometimes the poor estimator gets kicked around because his estimate doesn't line up to etiher the final estimated cost or the final actual cost.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[GregLamberson]

Fascinating discussion, of course the thread has been pretty much hijacked - sorry.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[BigInch]

Actually, at least on most of my projects, cost overruns have always been minimal. If I had an overrun, it was limited to engineering expenditures from engineering hours going over budget, which I never worried about because I have proven that for every 1 hour over budget in the engineering office meant 100 or more critical path construction hours saved in the field.

I'd say "most" cost overruns occur simply because of one of the following,

1.) Ill defined concept. The client keeps adding new and more beautiful things to the original scope of work and pfd so it'll go faster, fly higher, cary more crew or ammo, increased countermeasure equipment and in the end change the mission profile gets changed from only a fighter, to a fighter-bomber. etc. etc. etc.
2.) The work was spread out over a wide area and a number of contractors without sufficient project control manpower and/or underdeveloped project control systems and procedures.
3.) Corruption entered the picture.
4.) The "project" was to produce a complex computer program
5.) The DOD or DOS was involved (refer to item 3).
6.) Guerrilla actions,
6.) Excessive delays in final permitting and "finance" (refer to item 3).
7.) Poor design specification, project definition and construction specifications.
8.) Inadequate field investigations.

#7 being the least common of all.

http://virtualpipeline.spaces.msn.com

 

[MortenA]

Biginch

Have thought more of this i tend to agree with you (and Greg) now. I find that in major cost overrun (where i have some insight and thats not very often) especially 1) and 2) is a problem. What we as an engineering consultantcy often battle with is client that really dont know what they want.

Best regards
Morten