Rule of Thumb for estimating number of engineering deliverables 5

[PetroBob]

What are typical rule of thumb ratios that experienced engineers use when calculating number of deliverables for a process design? For example, if you know the approximate number of PFDs in a typical plant, how do you ratio to estimate of number of P&IDs, lines, major equipment, instruments etc? Recognizing that of course this will vary for each type of process. I'm looking for a starting point, say for a reasonably industrial process such as refinery, SAGD, LNG plant, Water Treatment, etc.

For example, how does the following look:

No of P&IDs = 6 x no of PFDs
No of LDT lines = 9 x no of P&IDs
No of plot plans = 0.5 x no of PFDs
No of major equipment items = 4 x no of PFDs
No of instruments = 7 x no of P&IDs

 

[pennpiper]

PetroBob,
You have the right idea here but you should make a spread sheet which allows for the different types of plants you noted in your post.

Example: A Crude Unit in a Refinery will need a higher multiplier (for the number of deliverables and hours per) than the Water Treatment Plant.

Request:
Please define "SAGD" & "LDT" for us.

prognosis: Lead or Lag

 

[MJCronin]

In order to calculate the number and type of process deliverables, you must develop preliminary PIDs and an equipment list.

You have not considered an important and expensive item in your estimation !.....the number of piping drawings !

The number of plot plans is usually pretty easy to estimate from job negotiations. How are the plot plans part of the process design ?

Where is the electrical estimate ? If the client decides to put his control room in the next county, the electrical costs can be stagering

I take HUGE EXCEPTION to your estimate of 7 instruments per PID.....I have worked on jobs where the number is closer to 100

I believe that each job is unique and you need a few geezers amongst the newbies to shepherd things along...

MJCronin
Sr. Process Engineer

 

[moltenmetal]

P&IDs arise from the equipment count, not from the PFD count. If you can't get a decent major equipment count from the PFDs, the PFDs are garbage. P&IDs, to be legible, need to have at most perhaps 2-3 pieces of equipment on them, and sometimes one truly major piece of equipment can have multiple P&IDs- then there are P&IDs for utilities distribution which have no equipment on them. Perhaps 1 P&ID per 1.5 pieces of equipment would give you a decent guess, but again that depends on how you do your P&IDs.

Instruments? Piping drawings? There's no meaningful way to estimate them from a PFD or equipment count. There's a huge variation based on project nature.

 

[MJCronin]

I fundamentally agree with MM..

Perhaps the best course of action is to ask for existing PIDs from the prospective client and make ratios and guesses.

I can't really see a client getting estimates from multiple parties (process engineering, piping, electrical etc) and then knitting it all together in the field.

Its best to have one consulting firm responsible for the WHOLE DOCUMENT PACKAGE and construction assistance.

Here is an excellent book on this subject:

http://www.amazon.com/Piping-Engineering-Leadership-Process-Projects/dp/0884153479

MJCronin
Sr. Process Engineer

 

[PetroBob]

Thanks folks for your wise advice. Apologies if this sounded like a dumb question. Asking here is one way to get smarter! Penpiper, SAGD is steam assisted gravity drainage, an oilsands / heavy oil in situ production process. LDT is line designation table, otherwise known as line list, which is associated with the P&IDs. MJCronin, thanks, good point that there are often a lot more than 7 instruments per P&ID. On reflection, yes, this seems very light for an average - and, as you say, will vary a lot depending on the process. It seems MM & MJ are making the point that the process needs to be developed sufficiently to get a major equipment count, before you can make this preliminary drawing count. This makes sense! And that P&ID count should be based on this rather than PFD count.
Thanks for the book reference - I have now bought a copy.
In this instance, we were looking at conceptual & front end design only, hence why only plot plan but not piping deliverables were being considered.

 

[MJCronin]

It is my experience that the piping material and installation costs can very strongly affect the overall cost estimate.

It will often be the factor controlling the final GO/NO-GO financial decision.

This is especially true regarding more exotic piping materials and valves.

MJCronin
Sr. Process Engineer

 

[car1615]

How would I go about estimating the number of valves that would be required for a 1 MPTA Ethylene cracker? Would I use same P&ID method? What would be the coefficient?

 

[moltenmetal]

You're kidding, right?

 

[car1615]

actually no, i'm looking at investment opportunities that come from the shale / ethane expansion in US, Estimated $50M spend in ethane related projects, ethylene, PP, PE, ammonia etc. My role is to look up the supply chain and determine what companies would benefit. one that I was looking into was the flow control space. This led me to better understand the amount of opportunity in pumps, instruments and valves. Thus the question about what the total spend would be for valves. I do understand the plant will be let out in pieces, but I am looking at the aggregate to see the total spend. 1-1.5% of capex seems likely, just trying to do some leg work here.

 

[car1615]

that should have stated $50B spend.

 

[moltenmetal]

Sorry, but there's no shortcut method that's going to give you a valve count within 1/2 of an order of magnitude.

 

[pennpiper]

car1615, You wrote:
"How would I go about estimating the number of valves that would be required for a 1 MPTA Ethylene cracker? Would I use same P&ID method? What would be the coefficient?"

You do not need to factor the number of valves. If you have the P&IDs than it is a simple matter to count the actual valves (by size and type) on the flow sheets and qualify the count as a +/- 25% estimate.



prognosis: Lead or Lag

 

[MJCronin]

Have we beat this to death yet ?

MJCronin
Sr. Process Engineer