Fast history matching and

uncertainty assessment
- Alternative history matched scenarios
- Reduce uncertainty
- Improve decisions
be dynamic

- Save time
 MEPO ®
Alternative scenarios

What is MEPO?
MEPO® is a step change in the reservoir engineering Today most projects end when a single history match is
workflow. It is the tool for assisted history matching. achieved. Project decisions are based on this solution, wells
MEPO works in harmony with your existing reservoir
®
are drilled, money is spent. But what if you have the answer 6
simulator and is simple to understand. It is capable x 6 and it should have been 4 x 9?
of history matching a reservoir much faster than
Using advanced mathematical techniques - such as
conventional manual history matching.
Evolutionary Strategies and Genetic Algorithms - MEPO® can
Save as much as 70% of the time you are spending find multiple solutions, quickly.
history matching today.
MEPO® can help.
MEPO® is able to find alternative solutions to the history
matching problem. Put simply, history matching is the
search for a mathematical solution that gives an answer to
a predefined number, or series of numbers you are seeking,
for example oil, water, or gas production, from a field or
well. Consider a very simple example; we need to create the
number 36 by multiplication, the solution could be 6 x 6, 4 x
9 or 12 x 3 they all give the answer we want, but which one
is correct?
MEPO ®
Reduce your cycle
time for production
forecasting

MEPO® is the solution for all oil companies challenged with

history matching, production forecasting and subsurface
uncertainty.

Reservoir simulation models are usually assumed to be

reliable if they reproduce the production history, but are there
other models that match the history, and do these models
produce the same predictions?

It has always been recognised that the best match does not
necessarily mean the best prediction, and that there will
always exist a range of good matches based on different
parameters. These different scenarios result in a spread of
future predictions, illustrating the inherent uncertainty of the
reservoir model and ultimately the volume of hydrocarbons Better yet, it can perform all this in a fraction of the time
to be produced. it would normally take to obtain a single history match by
conventional manual methods.
Until now it has simply been too time consuming to look for
alternative scenarios. MEPO® is now being used by an increasing number of oil
companies who are already reaping the rewards through
MEPO® enables simulation engineers to generate a represen- lower development costs, more efficient well placement, and
tative range of possible history matches, and assess these increased confidence in field development decisions.
based on quantitative measures of the match quality.
Are you working with a single realisation of your simulation?
Do you know the effect of alternative scenarios?
 Licences
Make more efficient
use of your simulation
licenses

How does MEPO® work? Efficient simulation license use

MEPO has an integrated pre- and postprocessor, which
®
MEPO® is able to work with “queuing systems” (e.g. LSF)
automatically checks the quality of the history match and to maximise the use of your simulation licenses. At busy
generates new input files for simulation. The user can interact times in the work cycle the number of simulation licenses
with the process at any time, allowing changes in history may be a bottleneck, but there are other times when you
matching strategy to be implemented with ease. could be extracting more value from your existing simulation
licenses.
The optimisation process in MEPO® works directly with your
simulator. There is no need to introduce response surfaces or MEPO® is able to use these quiet times efficiently to give you
specify times that you would like to match; however, you can a greater understanding of your reservoir. It is also possible
if you want to. MEPO is able to work with any parameter in a
®
to set a low priority on MEPO® generated runs and a higher
simulation deck. In fact the input and output is just the same priority on traditional runs so that the new technology need
as traditional history matching. not disturb existing work practices. Having a limited number
of simulation licenses is not a barrier to the MEPO® world.

Used simulation licenses

Unutilised simulation licenses
Scalability
Use a single CPU or a
multi node cluster

Simultaneous Simulations Supported simulators

Dependent on the business decision that needs to be taken MEPO® works with your existing reservoir simulator - we
you can scale the power of MEPO . It will work with a single
®
support most commercial reservoir simulators on Windows
simulation license, or MEPO can simultaneously launch as
®
XP, Linux or Sun Solaris platforms, and on hardware ranging
many simulations that you consider are needed to solve from PCs to Linux clusters.
your problem. This could be related to the time available
Eclipse© including Eclipse 100, Eclipse 300 (Schlumberger)
and the value of the decisions that needs to be taken.
FrontSim™ (Schlumberger)
Perhaps one, or two, or five, or eight, or more simultaneous
IMEX™, STARS™, GEM™ (CMG)
simulations. It is up to you.
VIP™ (Halliburton)
3DSL® (StreamSim Technologies)
Tempest™ - MORE™ (Roxar)
True scalability MoReS (Internal Shell Simulator)
CHEARS (Internal Chevron Simulator)
32
Proprietary simulators can be supported on request.
16
8
4
2

no. of CPUs/licences Time

 MEPO ®
A step change in
E&P workflow

Quickly determine the sensitivities in

your model
Before history matching a model, it is often useful to find
the sensitivity of the model to individual parameters.

MEPO® gives you the tools to do this efficiently. Experimental

Design Matrices (EDM) can easily be generated and used to
understand the effect of the uncertainty parameters. Pareto
plots and line plots help you quickly analyse the results to
establish the importance of a parameter. This will allow you
to effectively select the dominant uncertainty parameters
and eliminate unresponsive parameters from the history
matching process.

Such studies give you confidence that a history matching

solution exists within the range of parameters you have de-
fined. It is also a good method to find starting points for the
history matching process; alternatively you can use
an engineering best guess.
MEPO ®
Reduce your cycle
time for production
forecasting

Reducing uncertainty: improving reserve

estimates Wide uncertainty

EDM techniques are frequently used to obtain an understand-

ing of the possible range of hydrocarbons in place and the
range of recoverable hydrocarbons. This is a guide based
on the range of possible input parameters. There is no link
to production data. In an exploration scenario this is a very
useful guide.

However, in a field with production data it is possible to

reduce the level of uncertainty. The EDM range has many
combinations which, whilst theoretically possible, must be
rejected as they do not match the historical production data.

MEPO® is able to generate multiple scenarios (history Reduced uncertainty range

matched models). This collection of models provides a range
of hydrocarbons in place and a range of recoverable reserves
that agree with the production data and are therefore pos-
sible. This reduces the range and the subsequent uncertainty.
 Application areas
- History Matching
- Production Forecasting
- Development Planning
- Cycle Time Reduction
- Well Placement Optimisation
- Model Tuning
- Audit Trail

Make better field development choices

MEPO® is often used for ranking the locations of infill Different infill well locations ranked using MEPO®
wells, identifying locations which not only maximise addition-
al production, but also minimise risk. The ability to generate a
range of scenarios for production profiles based on alterna-
tive models reduces the risk of drilling an unproductive infill
well, which is more likely when only one predicted production
profile is considered.

Basing your choice of infill well location on a single

history match and production forecast can be dangerous.
MEPO® can help you understand the uncertainty range in
the production forecast and make smarter decisions.
Uncertainty
Are you over or under
forecasting future
production?
What are the
controlling factors?
MEPO ® will help

Quantify uncertainty in future production Do we see this in the real world?

Using MEPO , uncertainties in production can be
®
Take a look at the figure below looking at the range of
assessed by using alternative history matched scenarios, possible production from the same development plan but
all having acceptable matches to production history, but using differently history matched models.
based on different parameter combinations.
The range of possible outcomes is wide.

Are you drilling in areas that have a much higher risk than is
apparent today?

Or are you perhaps missing out on developments that have

Uncertainty in predictions potential?
 Case descriptions
MEPO® has been used on many
fields over the last four years. Most
projects are confidential. However,
some clients have gone to print.

On these pages you will find

information about some studies
that have applied MEPO®.

In all cases, many different

matches were found.

Enhanced field management in Siberia by quantifying production

TNK-BP uncertainties – SPE 101808
Tipping (TNK-BP), Deschenya (TNK-BP), Deimbacher and Kovyazin (Schlumberger).

West Siberia Conclusions

Lower Cretaceous Marine Automated-history matching techniques give major boosts to work
Sandstones and claystones productivity.
8 oil bearing beds
Field management decisions should account for uncertainties.
13 different oil water contacts
MEPO® – one week to HM, seven One history matched simulation model cannot be expected to give
different matches. accurate predictions of future production.
Traditional methods – nine months,
to obtain one match. Modern, cluster-based optimisation technologies offer massive reductions
in turn-around times for history matching and field development planning
Application of global optimisation methods for history matching and
probabilistic forecasting – Chevron case studies – SPE 105208
Choudhary (Chevron), Yoon (Chevron) and Ludvigsen (SPT Group).

Two West Africa Models Conclusions

Computer assisted history matching resulted in a three to four times speed
Model A up in history matching compared to the traditional approach.
25 years production; 2 wells
MEPO® - one and a half weeks to HM The quality of history matching is improved by using the assisted
Traditional methods - eight weeks approach.

Multiple solutions to the history match problem are identified.

Model B
26 wells; Sandstones and Shales Global optimisation methods are well suited to large numbers of
MEPO - four weeks to HM
® parameters.
Traditional methods – twelve weeks
The improved understanding of subsurface uncertainty captures upside
potential and reduces the downside risk of business decisions.

Application of global optimisation techniques for model validation and

OMV prediction scenarios for a North African oil field – SPE 100193
Griess (OMV), Diab and Schulze-Riegert (SPT Group).

North Africa Conclusions

The matches generate a range of different predicted production profiles.
30 years production data, 40 wells
MEPO® – four weeks to HM Dramatic improvement in the pressure match.

Traditional methods - twelve months. Modifications made on a layer basis across the whole model. No local
modifications used.

New Era of History Matching and Probabilistic Forecasting – A Case

Study – SPE 102349
Harneshaug (Statoil), Selberg, Ludvigsen, Diab (SPT Group).

North Sea Conclusions

Improved quality of match.
8 years of historical data
50 different scenarios found.
MEPO – two weeks
Traditional method – proved difficult Results show that a single history match model often fails to be
to match
representative of the possible production forecast range.
 SPT Group is a world leader in dynamic modelling
for the oil and gas industry. Employing highly
skilled professionals worldwide, SPT Group
provides a combination of software and
consulting services within multiphase
flow and reservoir engineering.

w ww.sptgroup.com

USA, Houston Tel: +1 281 496 9898 Norway, Oslo Tel: +47 64 84 45 50
Mexico, Mexico City Tel: +52 555 211 9211 Norway, Bergen Tel: +47 55 54 38 70
Australia, Perth Tel: +61 8 9286 6500 Germany, Hamburg Tel: +49 40 27 85 88 0
UAE, Dubai Tel: +971 4 426 4855 Russia, Moscow Tel: +7 495 135 88 16
Italy, Milan Tel: +971 50 453 9642 UK, London Tel: +44 1483 307 870
Malaysia, Kuala Lumpur Tel: +60 3 2161 4570

www.mepo.com [email protected]