More Production Optimization Discussion

Following on from the previous post, here are a couple more questions that I have tried to answer:

Do you think GAP is the best program for modelling 

From what I know, GAP is the most widely used of this type of program.  I think its main claim to being number one is that it works together with other good, comprehensive software, the Petex IPM suite.  The OPENSERVER functionality (not sure Santos has a license for that) was the thing that others did not have, basically you can write a function for anything you can do by hand and automate it.  So, I think you will find that other software may excel in certain areas and that is great if your field nicely fits with such a niche.  For example, I think LOWIS (Artificial Lift) and FastPIPER (Gas Fields) look like very good programs within their niches.   IPM still looks the best for general use.  

Should we be implementing systems to get engineers to look at data (e.g. THP vs meter run pressure or meter run pressure vs compressor suction pressure) to try and diagnose inefficiencies outside the model or do you think the main focus should be trying to make all of the parts of the model match reality and then investigating the bits that don’t?

Regarding the physical production systems, yes, of course engineers should be focused on optimising physical systems--that is the end goal.  I defined production system optimisation as the middle three steps of this process:

Data-->Surveillance-->Calibration-->Optimisation-->Execution

Should engineers look at their data sources and make sure they are functioning properly?  I have some trouble with that.  Of course engineers should be concerned about it.  However, the operations staff have the primary responsibility.  Engineers should set the tolerance parameters, should help Operations set up procedures, should select the right equipment, should justify upgrades and time spent, and should coach, explain and encourage Ops.  Yes, I would expect engineers to use the models (and common sense) to identify areas that justify improvements.

By the way, 'models' that we have been talking about include things like trends/plots.  Models are any tool that helps us know what to expect and what is unexpected.  Making a chart (or other visualisation) with expected ranges of data like meter run pressures and suction pressures that can spot anomalies is a worthwhile thing for engineers to do. Anomalies can be errors in data or physical problems/changes.  Spotting these can lead to hidden value.