See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/323994161

Extension of Dykstra-Parsons Model of Stratified-Reservoir Waterflood To

Include Advanced Well Completions

Article in SPE Reservoir Evaluation & Engineering · March 2018

DOI: 10.2118/189977-PA

CITATIONS READS

7 508

3 authors:

Khafiz Muradov Bona Prakasa Ritonga

Heriot-Watt University Heriot-Watt University
77 PUBLICATIONS 547 CITATIONS 8 PUBLICATIONS 55 CITATIONS

SEE PROFILE SEE PROFILE

David Davies
Heriot-Watt University
120 PUBLICATIONS 1,089 CITATIONS

SEE PROFILE

All content following this page was uploaded by Bona Prakasa Ritonga on 18 August 2020.

The user has requested enhancement of the downloaded file.

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 703 Total Pages: 16

Extension of Dykstra-Parsons Model of

Stratified-Reservoir Waterflood To Include
Advanced Well Completions
K. Muradov, B. Prakasa, and D. Davies, Heriot-Watt University

Summary
Large volumes of oil are being produced by waterflooding heterogeneous reservoirs. Careful flood-pattern design, well placement, and con-
trol are required to maximize oil recovery by delaying water breakthrough and optimizing sweep efficiency. Models that analyze the water-
flood’s performance and predict the production forecast, such as the Dykstra-Parsons (DP) method, are routinely used for this purpose.
The DP method estimates the vertical-flooding efficiency between conventional wells producing from noncommunicating layers.
This method and its various modifications have had a significant impact on the development of the theoretical description of the water-
flooding process. The DP method is still routinely used for waterflood-performance prediction and analysis, flood-pattern selection, and
recovery-factor calculation.
Advanced well completions (AWCs) control the fluid flow at the reservoir sandface. They have become a proven, widely used tech-
nology (particularly in waterflooded reservoirs) for modifying a production or injection well’s inflow/outflow rate profile along the
well. In addition to this, new AWC designs that react to water breakthrough have recently become available. Incorporating a description
of the AWC performance into the waterflood-analysis models will allow fast forecasting of the production profile and oil recovery, as
well as help in optimizing the AWC configuration and control at the well-design stage.
This work extends the DP method for rapid prediction of the waterflood’s performance to AWC wells. It provides a simple means of
estimating the additional, long-term value derived by controlling zonal flow rate using AWCs or any other means (e.g., well workover).
The accuracy of the extended DP method’s prediction has been verified by comparison against the results from a numerical reservoir
simulator. Several examples illustrate the extended DP model’s performance and value. The method’s limitations and possible future
extensions are also discussed.
The presented model is a simple, transparent approach to evaluating the impact on the waterflood’s oil recovery efficiency (RE) of
various well-completion and control options. It can be implemented as an analytical model or as a fast simulator. This model is also the
missing link between the various AWC design methods available today and the AWCs’ long-term value evaluation.

Introduction
Large volumes of oil are being produced, waterflooding heterogeneous reservoirs. Careful flood-pattern design, well placement, and
control are required to maximize oil recovery by delaying water breakthrough and optimizing sweep efficiency. Two important
approaches contribute to the successful waterflood design:
• Waterflood analysis and forecasting models, such as the DP method.
• Advanced well-construction designs, such as AWCs that control the fluid flow at the reservoir sandface. AWCs are a field-proven
technology that has been widely installed in thousands of wells to modify the production or injection well’s inflow/outflow rate
profile along the well. New AWC designs that additionally improve the waterflood’s performance by reacting to water break-
through have recently become available.
Incorporation of the AWC performance in a waterflood-analysis model will provide an accurate recovery-forecasting tool for
informed field design and operation in terms of the wells’ location, completion selection, and control to maximize oil recovery.
The AWCs subsection and the Simplified Methods for Waterflood Analysis subsection next will discuss AWCs, their history, design,
impact on well performance, and application envelope, followed by a description of the history and application envelope of the tradi-
tional waterflood-analysis methods.

AWCs. AWCs have flow-control devices (FCDs) installed in the production tubing in front of the production or injection intervals.
FCDs (one or several) can be installed as frequently as on every tubing joint, often amounting to hundreds per well. Annular flow isola-
tion (e.g., gravel pack or packers) is also preferred because it further improves the AWCs’ performance. Fig. 1 presents a schematic of
an AWC in an openhole production well with three packers, although AWCs can also be used in wells with a cased hole, wells with
multiple laterals, in injection wells, and others.
The operating principle of an AWC is simple: The pressure drop across an FCD is nonlinearly dependent of the fluid flow rate
(Eq. 1), unlike the pressure drop across the reservoir that is essentially linearly dependent on the rate. This ensures that there is an
increased pressure drop across the AWC for those inflow zones between adjacent packers producing with a higher production rate com-
pared with the less-productive zones. The result is a more uniform inflow/outflow profile along the length of the completion. FCDs can
be generally classified as passive (a fixed restriction), active (the restriction can be controlled), and autonomous (no control, but an
autonomous reaction to the presence of an unwanted fluid). A good overview of the evolution of AWC technology, the available types,
and their applications can be found in Eltaher et al. (2014) and Al-Khelaiwi et al. (2008).
AWCs are well-suited to wells producing from zones of differing reservoir quality, such as heterogeneous reservoirs, differentially
depleted layers, compartmentalized reservoirs, multiple-reservoir developments, unfavorably saturated (e.g., oil-rim) reservoirs, and
others. AWC well construction can be horizontal, deviated, or multilateral with the severity of the heterogeneity-related problems
increasing as the well/reservoir contact length increases. Passive AWC is also frequently installed to reduce the heel-to-toe effect in
homogeneous reservoirs developed with long horizontal wells.

Copyright V
C 2018 Society of Petroleum Engineers

Original SPE manuscript received for review 22 December 2016. Revised manuscript received for review 18 September 2017. Paper (SPE 189977) peer approved 14 November 2017.

August 2018 SPE Reservoir Evaluation & Engineering 703

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 704 Total Pages: 16

Production casing
Production packer

Annular flow isolation

Production tubing

Sandface (wellbore wall)

Flow-control device

Fig. 1—Schematic of a well with AWC.

In fact, passive FCD technology began with the development of the inflow-control device (ICD) in the early 1990s with the aim of
controlling the inflow profile to minimize the heel-to-toe effect in long horizontal wells producing from the Troll field, a massive,
homogeneous, oil-rim reservoir offshore Norway (Henriksen et al. 2006). The ICD market has expanded since then [e.g., Baker-Hughes
had installed more than 2-million ft of helical channel ICDs by 2008, Statoil has more than 120 North Sea wells with ICD completions,
and Saudi Aramco has 200 ICD completions in several fields (Al-Khelaiwi 2013)].
Passive AWC. This is composed of a range of commercial ICD types of differing design and tolerance to erosion and their response
to the properties of the flowing fluid. However, all ICDs are essentially a fixed restriction with a performance that can be described by

DPICD ¼ aQ2 ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð1Þ

in which the “strength” parameter a relates the extra pressure drop DPICD added by an ICD restriction to the square of the flow rate Q.
More information on ICD types and the formulae to find the strength a for each ICD type can be found in Al-Khelaiwi (2013).
Active FCDs. Often called interval-control valves (ICVs), they vary in design depending on their source of power and control signal
(electric or hydraulic), how many positions can be chosen (usually between 2 and 10), and others. Eq. 1 also describes the pressure drop
across an ICV for a given position. The well’s production is frequently controlled with the ICV in a reactive manner. Initially, all ICVs are
set to the fully open position, and then the selected ICVs are shut after onset of high water production in the zones they control. The alter-
native, proactive control, involves setting the ICVs to different restrictive positions from the very beginning to manage the well inflow
and the water-propagation front to maximize the long-term oil recovery. ICVs may also be managed to increase production by improving
the tubing outflow performance. Finally, the ICVs may be used instead of a well-workover operation/intervention, a particularly popular
option for subsea wells. More information on ICV types, their applications, and control can be found in Haghighat Sefat et al. (2016).
The recently introduced autonomous inflow-control device (AICD) and valve (AICV) react to “unwanted” (in oil production) fluid
phases (i.e., free gas and water), restricting their flow in situ and improving recovery. The AICD/AICV performance is described sepa-
rately for single-phase oil flow and for single-phase water or gas flows, because these performances differ significantly. However, in
both cases, their single-phase-flow performance can be acceptably matched by Eq. 1. (The AICD/AICV multiphase-flow performance
is still difficult to model accurately. The multiphase-modeling problem is avoided in this work by only considering single-phase flow
situations, as explained next.) More details about the AICDs/AICVs can be found in Eltaher et al. (2014). For simplicity, we will refer
to both the AICDs and AICVs as AFCDs (autonomous flow-control devices) where their distinction is not explicitly required in the
next sections.
Maximum value added by the AWC is achieved only if the AWC is properly designed. The AWC-design process requires specifica-
tion of the FCD types, the restriction levels, and the number of devices as well as the specification of the annular flow isolation. AWC
design is a complex mathematical problem that, so far, has been approached in different ways.

The Value Added by AWC to the Whole Field Production Is Evaluated. This is necessary because AWC action has a long-term
impact on the field recovery. This ideal approach to the AWC design implies a multidimensional optimization problem using a numeri-
cal reservoir-simulation tool to forecast the added value for an arbitrary AWC configuration. This optimization problem is challenging
and, if carried out, is approached sequentially. For example, the well placement is optimized first, then the annular flow isolation for an
arbitrary AWC (MoradiDowlatabad et al. 2014), and finally, the AWC design is optimized. No examples of such integrated optimiza-
tion, to our knowledge, have been published to date.

The Standard (Snapshot Design) Approach Using a Wellbore Model Is To Design the AWC To Improve, but Not Optimize, the
Oil Recovery. This simplification uses a stand-alone wellbore simulator, an approach which is much faster than the use of a fully
coupled wellbore-reservoir simulator, as described previously. The reservoir inflow/outflow performance is estimated at a specific time
and input into the wellbore model. The start of well production is often chosen as the specific time. The wellbore model is then used to
vary the AWC specifications, thereby reducing the flow imbalance between the zones while avoiding an overwhelmingly great restric-
tion to the inflow. The result is a more uniform waterflood front, believing that this will improve the oil recovery. However, this
assumption is not normally rigorously tested with a full reservoir simulation. A range of such snapshot methods has been developed.
Birchenko et al. (2011) developed analytical and semianalytical methods to reduce both the heel-to-toe effect in a homogeneous reser-
voir and the heterogeneous inflow imbalance (Birchenko et al. 2011). Prakasa et al. (2015) used a type-curve method for ICD comple-
tion design, whereas numerical wellbore model-based designs (Al-Khelaiwi 2013) and inflow/outflow balance methods (Al-Khelaiwi
2013) have been proposed.
The previous snapshot methods lack the ability to predict the AWC design’s long-term impact on the development of the waterflood,
the oil RE, fractional flow-rate profile, and others.
This paper offers a solution to the problem of predicting the long-term impact of AWC on the waterflood performance.

704 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 705 Total Pages: 16

Simplified Methods for Waterflood Analysis. Waterflood is the most commonly used oil-recovery mechanism. Analysis of its per-
formance forecast has been a popular research topic since the 1940s. Simplified analysis methods offer (semi-) analytical solutions and
type curves used to understand and to forecast waterflood performance relatively well in a range of geological environments, flood pat-
terns, and well-control situations. Simplified analysis methods are still routinely used to evaluate the expected oil recovery, waterflood
efficiency, efficiency of the flood pattern, and others despite the widespread availability of numerical reservoir simulators.
The oil-recovery factor is the product of the efficiency of oil displacement by water (at the pore level) and the volumetric sweep effi-
ciency (the extent of the water front at the reservoir level). The latter is the product of the areal and vertical sweep efficiencies. The
areal sweep efficiency is very much the function of the flood pattern, geological discontinuities, mobility ratio, and others. The vertical
sweep efficiency at any given time is the fraction of the cross section between the injector and the producer that is waterflooded. This
efficiency strongly depends on the reservoir heterogeneity—more-permeable layers flood faster, distorting the water front.
Prediction and analysis of the vertical sweep efficiency are presented in this paper because AWCs can deal with uneven flood-front
propagation as encountered in heterogeneous reservoirs. The areal-sweep and oil-displacement efficiencies can be derived separately
from, for instance, mathematical models (Dyes et al. 1954) and coreflooding experiments, respectively.
Analytical solutions for vertical sweep efficiency in heterogeneous, stratified reservoirs can only be found after making certain
assumptions, including the one on the level of interlayer communication in which two different situations can be assumed: The layers
are either noncommunicating or are perfectly communicating with instantaneous, vertical pressure equilibrium.
The case of “noncommunicating layers” with arbitrary properties of layers was well-analyzed by Dykstra and Parsons (1950). They
derived the analytical solution (called here, the “DP method”) to be applied on a layer-by-layer basis for estimation of the RE, fractional
flow curves, and others with respect to the injected water volume. They also used these analytical solutions to derive the general formu-
lae and type curves for the reservoir with log-normal, vertically distributed permeability. Later contributions include work by Muscat
(1950), who derived expressions for other types of permeability distribution, whereas the situations with multiple water/oil ratios were
analyzed by Johnson (1956). The model prediction with respect to injected volumes was translated into the time domain by Reznik
et al. (1984) and El-Khatib (1985). Osman and Tiab (1981) extended the model to the case of composite layers (i.e., in which perme-
ability varies laterally). Nonpiston-like displacement was considered by Snyder and Ramey (1967), who added elements of the
Buckley-Leverett theory.
Warren (1964) and Goddin et al. (1966) considered the effect of crossflow between layers. The waterflood-analysis solutions in
“communicating layers” were presented by Hiatt (1958) with later, major extensions to the cases of variable layer properties (El-Khatib
1985); reservoirs with log-normally, vertically distributed permeability (El-Khatib 1999); and even included the gravity-caused cross-
flow effect (El-Khatib 2003), including inclined reservoirs (El-Khatib 2010).
Our work extended the DP method for stratified reservoirs with noncommunicating layers to the case of wells with AWCs. A simple
and quick approach to predict waterflood performance is offered, allowing estimation of the AWCs’ long-term, added value. We veri-
fied our method against a numerical-simulation model prediction, but also discussed why a similar approach is unlikely (1) for commu-
nicating layers or (2) to include permeability distributions.
We first briefly describe the original DP method, followed by our extension to include the performance of an AWC. The algorithm
to implement the extended DP method workflow is fully described. We then compare the method’s performance against a numerical
reservoir simulation using a synthetic reservoir model, and illustrate how the method can be used to design and evaluate the long-term
added value of an AWC.

DP Method and Its Extension To Incorporate AWC Performance

All equations are in SI units unless otherwise stated with all volumes, and rates are measured at reservoir conditions. Nomenclature,
terms, and abbreviations are tabulated at the end of the paper.

DP Method for Simplified Analysis of Oil Displacement by Water in a Noncommunicating Layered Reservoir. The DP method
is traditionally used for vertical wells, but can also be used for predicting the performance of a waterflood with deviated or horizontal
wells (with tilted layers or channels) when the assumptions mentioned later are satisfied. This justifies our extension of this method to
AWC wells that are not always vertical.
Note that the equivalent methods for a communicating layered reservoir do exist. However, they assume perfect vertical pressure
equilibrium for all layers (i.e., the local pressure derivative at a given distance from the well is the same for every layer). Unfortu-
nately, they do not apply to wells with AWCs because the AWCs are intended to create significantly different pressure drops across
each layer.
The DP method (Dykstra and Parsons 1950) and its modifications are still extensively used to estimate the reservoir’s vertical sweep
efficiency, an integral part of estimating the recovery factor. This original DP method for conventional wells without AWC will now be
described by considering a heterogeneous reservoir with an injector/producer pair separated by distance L (Fig. 2).
Each layer has its own properties: height h, effective
 cross-sectional
 area A, porosity U, horizontal permeability
 k, endpoint satura-
tions Swi and Sor, endpoint mobilities k0w  krw 0
lw and k0o  kro
0
lo , and endpoint mobility ratio M0  k0w k0o . The effective cross-
sectional area is the area that is being flooded [i.e., this area (in a given layer) multiplied by the water flow velocity in a layer should
give the injection flow rate for this layer]. [Note that if the flood of tilted sands by a horizontal injection/production well-pair is ana-
lyzed, unlike the vertical-wells example we discuss here, then, the area A can be found using the same principle, and is perpendicular to
the (lateral) flood direction, across the tilt]. We assume the following.
Noncommunicating Layers. This is normally expressed by the stronger statement kv ¼ 0. This stresses the fact that the gravity
forces’ layer-by-layer impact on the oil-displacement front is minor compared with the overall lateral flow. This is true for most hetero-
geneous reservoirs.
Piston Displacement in Each Layer. This expresses the fact that the lateral water-front variation within a given layer is minor com-
pared with the overall front variation caused by the reservoir heterogeneity. This is also true for most heterogeneous reservoirs.
A Constant Pressure Drop Between the Injection and Production Wells (or, More Specifically, Between Tubing-Intake
Pressures). This is reasonable because the injection and production wells are normally controlled by pressure for most of their produc-
tion life. Similarly, 100% voidage replacement, with each layer’s injection rate matching the production rate, is assumed. Liquids and
rocks are modeled as incompressible. The Workflow for Simplified Analysis of Oil Displacement by Water in a Noncommunicating
Layered Reservoir Developed by Wells With FCC subsection shows how our method can be extended to constant production rate.
Fig. 3 considers a Layer j where the water front has traveled to coordinate xj.

August 2018 SPE Reservoir Evaluation & Engineering 705

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 706 Total Pages: 16

Injector Producer

Pinj ΔP = Pinj – Pprod Pprod

Water Oil
Layer N

Layer j

Layer 2

Layer 1

Sandface flow-control completion

Fig. 2—Schematic of oil displacement from an injector to a producer at some time in a heterogeneous reservoir.

Layer j
Xj L – xj
Water front

So = Sor,j Sw = Swi,j

ΔPw,j ΔPo,j

ΔP = Pinj – Pprod

Fig. 3—Saturation profile in Layer j.

Darcy’s equation states that the flow velocities in oil and water regions are, respectively,
kk0o DPo
uo ¼  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð2Þ
Lx
kk0 DPw
uw ¼  w : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð3Þ
x
Fluids are assumed incompressible, so these velocities are equal to each other; that is, uo ¼ uw ¼ u. dx
By noticing that DP ¼ DPoþDPw and by combining this equation with Eqs. 2 and 3, the front velocity [equal to the flow velocity
u divided by the movable pore volume (PV) fraction] for Layer j can be found as: dt

dxj uj kj k0w; j DP
¼ ¼  ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð4Þ
dt DSj /j DSj /j L xj þ Mj ð1  xj Þ

where we have introduced dimensionless front distance x* defined as xj  xj =L , and the movable saturation DS defined as DS ¼ 1 
Swi  Sor.
Now, clearly, the production or injection rate q is a function of the water-front distance:
kj k0w; j Aj DP
qj ¼ uj Aj ¼   : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð5Þ
L xj þ Mj ð1  xj Þ

The time-independent oil RE vs. injected volume curves for oil-displacement analysis can be constructed by eliminating time from
the equations and focusing on relative front positions. The ratio of the front velocities in Layer j and in the (Reference) Layer R is
dx,
j
dxi x þ MR ð1  xR Þ
dt ¼ ¼ Fj;R R ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð6Þ
dxR dxR xj þ Mj ð1  xj Þ
dt
kj k0w; j DSR /R
where Fj,R is defined as Fj;R   .
DSj /j kR k0 w;R
706 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 707 Total Pages: 16

The original DP derivation work flow then integrates Eq. 6,

xj
ðh i ð1
 
xj þ Mj ð1  xj Þ dxj ¼ Fj;R xR þ MR ð1  xR Þ dxR ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð7Þ
0 0

to find an analytical solution for the water-front position x in Layer j at the exact time when Layer R experiences water breakthrough
(i.e., Layer R is flooded when xR ¼ 1). Further, the integral solution can be expressed explicitly for xj as
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Mj  Mj2 þ Fj;R ð1  Mj Þð1 þ MR Þ
xj ¼ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð8Þ
 Mj  1 
Fj;R ð1 þ MR Þ
except for the case Mj ¼ 1 when the integral solution yields xj ¼ .
2
[It is also possible to find the imaginary distance (and hence, the injected volume) of the water front in a layer, after it has been
flooded, by changing the integration limits.]
In this way, the front distances in the layers are calculated, taking the first layer to experience breakthrough as a reference layer. This
allows calculation of both the overall reservoir saturation and the injected volumes at this time. The production rates of oil and (in the ref-
erence layer) water are calculated with Eq. 5. This allows the calculation of the production well’s fractional flow fw [or watercut (WC)].
Repeating this step for each subsequent layer in the order they experience water breakthrough, the curve of reservoir water saturation
or injected volume vs. WC is constructed. This allows analyzing the waterflood’s dynamics, efficiency, and recovery-factor estimation.
Thus, the main objective of the method is achieved.
[If the layers are differentially depleted, their permeabilities can be proportionally corrected for this as proposed in Dake (2001).]
Optionally, time also can be reintroduced into the model. This may be important when the actual production/injection forecast is
needed (e.g., for evaluating the field economics). Time can be introduced accurately by integrating the analytical solution for xj (Reznik
et al. 1984; El-Khatib 1985). Alternatively (or without an alternative in the case of AWCs as will be seen in the Work flow for Simpli-
fied Analysis of Oil Displacement by Water in a Noncommunicating Layered Reservoir Developed by Wells  With FCC section), time
WIR 
can be approximately introduced by matching the injection rates and injected volumes; that is, tbt;R  . “Approximately” is
qR at x ¼1
R
used here because the injection rate changes with time (yet, not significantly) until water breakthrough.
The Extension of the DP Method to the Case of Wells With AWC section will follow the same work flow while introducing the
extended solutions able to incorporate the impact of an arbitrary sandface flow-control completion (FCC) in wells.

Extension of the DP Method to the Case of Wells With AWC. Similar to the pressure drop across a single FCD (Eq. 1), for a given
flowing phase, the sandface FCC also imposes a pressure drop quadratically proportional to the flow rate:

DPFCC ¼ aflowing phase q2 : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð9Þ

The coefficient a (FCC strength) is a function of the FCD strength, the number of FCDs per zone in question, and the flowing-fluid
properties. Various formulae explaining how to calculate it for a given FCD type or for a completion zone containing multiple FCDs
are presented in Al-Khelaiwi (2013). In this study, for simplicity, we will refer to it as aw for water flow and ao for oil flow.
The total pressure drop across a layer is now a function of both the reservoir and FCC pressure drops:
DP ¼ DPFCC;injector þ DPlayer þ DPFCC;producer ¼ ðainj þ aprod Þq2 þ DPlayer ¼ a q2 þ DPlayer ; . . . . . . . . . . . . . . . . . . . . . . ð10Þ

where a for a given layer equals a  abbt ¼ aw;inj þ ao;prod before water breaks through this layer; and a  aabt ¼ aw;inj þ aw;prod after
water breakthrough. It is possible to model any arbitrary FCC performance or well-workover action for any given layer by changing
coefficients a for this layer. The following FCC examples illustrate what can be modeled.
• ICD, ICV, AICD, AICV completion (with any number of devices per layer) when the a coefficients are selected to match the
FCD’s number/layer and single phases’ performance curves.
• a ¼ 0 for a conventional well with no FCC.
• Either aw,prod ¼ 1 or qabt ¼ 0 represents closure of a zone by a well workover, an ICV, or autonomous water shut-in response of
the FCC.
• Any other arbitrary flow-control action for each layer.
This allows modeling any FCC performance.
The original DP method is valid for the reservoir-only flow (i.e., when the impact of FCC is eliminated and sandface pressures are
considered). To find this sandface-to-sandface pressure drop across any layer DPlayer, Eq. 10 is rearranged as
DPlayer ¼ DP  a q2 : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð11Þ
kj k0w; j Aj DP  aj q2j
Changing DP to DPlayer in Eq. 5 results in qj ¼   , and expressing this quadratic equation for u ¼ q/A yields
L xj þ Mj ð1  xj Þ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Bj þ Bj 2 þ 4aj A2j C2j DP
uj ¼ ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð12Þ
2aj A2j Cj

where coefficients B and C are defined as

Bj  xj þ Mj ð1  xj Þ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð13Þ
kj k0w; j
Cj  : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð14Þ
L

August 2018 SPE Reservoir Evaluation & Engineering 707

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 708 Total Pages: 16

Note that the effective cross-sectional area is introduced immediately in the AWC case because it affects the reservoir pressure
drop. This is different from the original DP method in which this area appears later when calculating the flow rates and water cuts. This
demands the effective cross-sectional area to be estimated reliably, taking into account, for example, the areal sweep efficiency, flood
patterns, faults, and others.
Following the same logic that led to Eq. 7 in the original method, we solve the following integral:
xj xðR
ð
DSj /j  DSR /R 
dxj ¼ dxR : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð15Þ
uj uR
0 0

The general analytic solution for Eq. 15 for arbitrary limits is

xj
2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3xj
ð B 2 þ 4a A2 C2 DP þ B
Bj
DSj /j  DSj /j 6 j j j j j
7
dxj ¼ 4 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 5 : . . . . . . . . . . . . . . . . . . . . . . . . . . ð16Þ
uj 4ð1  Mj ÞCj DP þ 4a A2 C2 DPln B 2 þ 4a  A 2 C 2 DP þ B
xj;min j j j j j j j j  xj;min

Specific limits are considered in the Work Flow for Simplified Analysis of Oil Displacement by Water in a Noncommunicating Lay-
ered Reservoir Developed by Wells With FCC subsection, where, finally, the complete application workflow is presented.

Work Flow for Simplified Analysis of Oil Displacement by Water in a Noncommunicating Layered Reservoir Developed by
Wells With FCC. Consider an injector/producer pair, separated by distance L, in a heterogeneous reservoir (Fig. 2). Each layer has its
own height  h, effective cross-sectional
 area A, porosity U, horizontal permeability k, endpoint saturation Swi and Sor, endpoint mobilities
k0w  krw0
lw and k0o  kro
0
lo , and mobility ratio M  kw =ko . Injector and producer can have FCC, with an impact on each layer’s flow
that is described by the a* coefficient defined by Eq. 9. Note that a* can vary (e.g., abbt before breakthrough and abbt after breakthrough).
The next steps, following the original DP method while using the equations derived in the Extension of the DP Method to the Case
of Wells With AWC subsection, describe how the waterflood analysis can be performed.
Step 1. Order the layers by breakthrough time, approximately estimated as the interwell distance divided by the front velocity calcu-
lated at the time when xj ¼ 0.5, derived using Eqs. 12, 13, and 14:

, 8 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi91
DS / <½0:5 þ Mj ð1  0:5Þ þ ½0:5 þ Mj ð1  0:5Þ2 þ 4aj;bbt A2j C2j DP=
 L dx 
j j
tbt; j ¼L ¼ LDSj /j : . . . . ð17Þ
j uj ðxj ¼ 0:5Þ : 2aj;bbt A2j Cj ;

dt xj ¼0:5L

Step 2. Take the first layer to experience breakthrough as the reference layer. Find the other layers’ front positions when the reference
layer experiences water breakthrough (i.e., when xR ¼ 1). This is performed by equating the solutions described by Eq. 16 as follows:
xj
ð ð1
DSj /j  DSR /R 
dxj ¼ dxR ;
uj uR
0 0

which yields
0 nqffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi o qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1
ðxj þ Mj ð1  xj ÞÞ ½xj þ Mj ð1  xj Þ2 þ 4aj;bbt A2j C2j DP þ ½xj þ Mj ð1  xj Þ  Mj Mj2 þ 4aj;bbt A2j C2j DP þ Mj
B C
B C
DSj /j B 8q ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 9 C
B > > C
B
ð1  Mj ÞCj B <   2  2
½xj þ Mj ð1  xj Þ þ 4aj;bbt Aj Cj DP þ ½xj þ Mj ð1  xj Þ 2   = C
 C
@ þ 4aj;bbt Aj Cj DPln
2 2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi A
>
: 2
M þ 4a A C DP þ M 2 2 >
;
j j;bbt j j j

2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3
1 þ 4aR;bbt A2j C2R DP þ 1Þ  MR ð MR2 þ 4aR;bbt A2R C2R DP þ MR
6 7
6 7
DSR /R 6 6 0 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 7
7
¼
ð1  MR ÞCR 6
6 1 þ 4a  A 2 C2 DP þ 1 7
7              ð18Þ
4 þ 4aR;bbt A2R C2R DPlnB C
R;bbt R R
@qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi A 5
2  2
MR þ 4aR;bbt AR CR DP þ MR2

To test the derivation validity at this stage, one can set the FCC coefficients a in Eq. 18 to zero and to check that the result would
then match the original, no-FCC solution described by Eq. 8.
Note that xj cannot be expressed explicitly as in the original method (Eq. 8), and instead should be solved numerically. Modern pro-
grams (e.g., MS Excel or Matlab) allow solving such equations in a fraction of a second.
Step 3. These front positions can be used to calculate the injected water volume in cubic meters (WI) or in reservoir PVs (WI*), as
well as the reservoir oil RE as follows:
X
N
WI ¼ Aj DSj /j xj L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð19Þ
j¼1

708 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 709 Total Pages: 16

X
N
Aj DSj /j xj L
j¼1
WI  ¼ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð20Þ
X
N
A j /j L
j¼1

X
N 

Aj DSwi; j /j xj L

j¼1 
RE ¼ N  : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð21Þ
X 

Aj /j Lð1  Swi; j Þ

j¼1 if xj 1 then xj ¼1

Note that xj in Eq. 21 equals 1 if xj >¼ 1 (see also Step 2a to calculate xj for the layers that had previously experienced breakthrough in
the case of subsequent reference layers).
Step 4. The production flow rates for each layer at the time of water breakthrough into Layer R are calculated as qj ¼ uj Aj , where uj
is calculated with Eqs. 12, 13, and 14. The reference layer is producing water while the other layers are still producing oil. Then, the
fractional flow rate of water fw is calculated as
X N  
1 if xj  1
qj 
0 if xj < 1
j¼1
fw ¼ N   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð22Þ
X 1 if xj  1
qj  
1 if xj < 1
j¼1

[If WC is preferred instead of fraction flow, “1 if xj < 1” should be changed to “1/bo if xj < 1”, where bo is the oil formation volume
factor (FVF)].
Step 5. Select the next layer to experience water breakthrough as the reference layer, and repeat Steps 2 to 4. Continue until all
layers have experienced breakthrough.
Step 2a. For a given reference layer, the layers that had experienced water breakthrough earlier have been producing water for
some time. The original method describes how this can be considered when calculating WI. Namely, an imaginary water-front position,
propagating beyond x* ¼ 1, is calculated and used to calculate the volume of water injected into an already flooded layer. Assume the
reference Layer R is experiencing water breakthrough, whereas Layer k had experienced breakthrough earlier and has been producing
water. To calculate the relative front positions in this case, use

ð1  xk
ð  ð1
DSk /k   DSk /k   DSR /R 
dxk  þ dxk  ¼ dxR : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð23Þ
uk a¼a  u k a¼a uR
bbt abt
0 1 0

The integral’s solution is given in Eq. 18, where the integral limits should be taken from Eq. 23.
The second integral on the left-hand side in Eq. 23 describes the imaginary propagation of the water front beyond the production
well. Layer k, having experienced water breakthrough, is producing water. The production well’s FCC across Layer k may have
changed its strength as it responded to water during breakthrough, so the appropriate aw strength should be used in the second integral.
Note: This step is omitted for Layer k; in other words, no further water injection into the layer is assumed, if the zone (between two
packers) is shut-in upon water breakthrough as could occur with reactive control by an ICV, a water-restrictive AICD, closing a sliding
sleeve, well-workover operation to abandon that zone, and others.
The example of this work flow application, and the acceptable match of the results with a comprehensive, numerical reservoir-simu-
lation model prediction, are shown in the next section. This work flow was quickly realized in an Excel spreadsheet, and is solved in a
fraction of a second for arbitrary boundary conditions and FCC configurations.
Finally, (A) time can be introduced into the production forecast, as well as (B) the method can be modified to model the constant
production-rate cases (as opposed to the constant pressure drop between tubing intakes as assumed in the previous calculations):
A. Introducing Time Into the Extended DP Method. The traditional, accurate methods of translating the injected volume into time by
integrating the injection rate over front position x are not applicable here, because there is no explicit analytical solution for x.
Eq. 17 is used to find the approximate time of each breakthrough, and thus, to translate the calculated injected volumes, oil RE,
fractional flow, or WC, and others into the time domain. Naturally, the flow velocity changes as the water front propagates within
the layer (see e.g., Eq. 2), but this change is not that significant; thus, the velocity estimate at x* ¼ 0.5 is sufficient. This accuracy
is confirmed in the Model Verification and Example Applications section.
B. Extending to the Constant-Rate Case. The constant pressure-drop workflow easily can be extended to the constant-rate case. For
an arbitrary pressure drop, the approximate layer rates are estimated by calculating the flow rates at the time when xj ¼ 0:5 using
Eqs. 12, 13, and 14. Their sum gives the total well rate. An optimizer is then used to find the pressure drop that gives the required
well-production rate. This pressure drop is then used as explained in the workflow described previously.
Overall, this adds a small extra calculation when the model is set up.
Note that the well’s approximate flow rate used to match the reservoir pressure drop was estimated before water breakthrough in the
“constant-rate case.” The flow rate may change significantly at a given pressure drop if there is a strong FCC action after-breakthrough
(e.g., zonal shut-in). In this case, the integration interval must be split into smaller intervals related to the periods before and after the
zonal action with the appropriate changes made to the rates, pressure drops, and FCC coefficients.

Model Verification and Example Applications

Model Verification. Oil recovery and fractional flow (or well WC) predicted by our extended DP model has been compared with the
results from a numerical reservoir simulator. A box-shaped reservoir model with dimensions of 10 500 30 m, was built with 5 non-
commutating layers (see Table 1 for the modeled properties). The distance, L, is 500 m between the vertical injection and production

August 2018 SPE Reservoir Evaluation & Engineering 709

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 710 Total Pages: 16

wells on opposite sides of the reservoir. (The modeled width, 10 m, is small compared with the well-spacing (500 m). It was chosen to
ensure the areal sweep efficiency is 100%, because only the vertical sweep efficiency is considered and the reservoir cross-section
waterflood is essentially modeled).

Layer k (md) h (m) ϕ µw (cp) µo (cp) Swi Sor kw Endpoint ko Endpoint Area Perpendicular to Flow (m2)

1(top) 5,000 5 0.45 1 1.5 0.2 0.1 0.5 1 50

2 1,000 7 0.25 1 1.5 0.2 0.25 0.4 1 70
3 300 9 0.2 1 1.5 0.2 0.4 0.6 1 90
4 2,000 6 0.3 1 1.5 0.2 0.3 0.7 1 60
5 4,000 3 0.4 1 1.5 0.2 0.15 0.5 1 30

Table 1—Layer properties in the test reservoir model.

The wells, controlled by bottomhole pressure (BHP), kept a constant (tubing) BHP difference between the injection and production
wells of 20 bar. The relative permeability curves used in this model are straight lines connecting the endpoint values listed in Table 1.
[In general, the analytical solution derived previously is meant to be applicable to any shape of relative permeability curves when the
stated conditions apply: (a) the minimum oil (Sor) and water (Swc) saturation can be achieved; (b) the flood-front spread within a layer is
smaller than within the whole reservoir (i.e., the “piston-like displacement” assumption). The straight-line relative permeability curves
are used in this study because they satisfy these conditions well]. The simulation time was chosen to be sufficiently long to ensure that a
sufficient water volume (approximately 2 PV) was injected to fully flood (i.e., to fully recover mobile oil) the reservoir. Table 2 lists
the well-completion options modeled.

Layer FCC Strength a*Before Breakthrough Layer FCC Strength a*After Breakthrough
2 2
Case Name [bar/(rcm/d) ] [bar/(rcm/d) ]

a* 0 0 0 (openhole completion) 0 (openhole completion)

a* 0.008 0.008 0.008 (medium-strength FCC) 0.008 (medium-strength FCC)

a* 0.016 0.016 0.016 (high-strength FCC) 0.016 (high-strength FCC)

Table 2—Well-completion options modeled in the test case.

The number and strength of the FCD restrictions in the reservoir model were chosen so that the layer FCC strength added a reasona-
ble pressure drop across the completion in front of each layer. (Note that “reasonable strength” in ICD technology is understood as the
strength causing the FCC pressure drop to be comparable to the drawdown.) The Table 2 values for FCC restriction levels are realistic
and resemble those encountered during the operation of real fields.
The zonal FCC strength, a*, represents the combined performance of the FCCs in both the injection and production wells for a given
layer (also see the comments on Eq. 10). Note that the Table 2 values of a* do not change after water breakthrough, despite water being
produced from the production well. The values in Table 2 thus represent a fluid-property-tolerant completion (e.g., “hybrid” ICDs).
Also, note that the FCC was designed to have the equal FCC strength for each layer. This allows robust model verification by simplify-
ing the comparison and analysis of our model’s results with those from the numerical reservoir simulation.
The cases are named as “[a* (or FCC strength for every layer)] [strength before breakthrough in bar/(rcm/d)2] [strength after break-
through in bar/(rcm/d)2]”; for example, Case “a* 0.008 0.008” implies that the combined FCC strength for each layer before and after
water breakthrough is 0.008 bar/(rcm/d)2. The conventional-wells case (that can also be modeled by the original DP method), or base
case with no FCC, is named “a* 0 0”.
Fig. 4 shows the RE vs. injected-water volume measured in reservoir PVs for both the numerical simulation and our semi-analytical
model. The two models match almost perfectly. We therefore consider the extended DP model to have been verified.
Note that, as expected, the stronger FCC achieves a higher recovery at a given injected volume. The more-productive layers are
restricted more strongly, and the water-propagation front is more uniform, resulting in a better sweep (Figs. 5 and 6). Also note that the
effect is nonlinear. There is a clear gain in RE after the FCC strength is increased from 0 to 0.008 bar/(rcm/d)2, but the gain caused by
the change of strength from 0.008 to 0.016 bar/(rcm/d)2 is less pronounced.
The next subsection presents further verification of the extended DP model. Two examples, (1) an ICD completion and (2) an
AFCD completion, illustrate the value of the extended DP model.

Examples of Application of the Extended DP Method. We will now illustrate the value that can be derived from the extended DP
method for modeling and designing AWCs. We continue verification of the semianalytical, extended DP model by comparing the
results with those from the same, Model Verification section, numerical reservoir-simulation model.
ICD Completion-Design Example. The objective of this example is to find the FCC strength a*, thereby improving oil recovery. A
reservoir numerical simulation is the best tool to evaluate the different FCC design options and their long-term consequences, as
explained in the Introduction section. By contrast, the most commonly used tool, which is much less comprehensive but very simple to
use, is a standalone wellbore model that is only capable of identifying the trade-off between balancing the zonal inflow and the well-
productivity loss at the start of well production. The approach is simple, fast to implement, but misses out on the interaction with the
reservoir and the other wells, and fails to evaluate any long-term consequences of an AWC design.

710 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 711 Total Pages: 16

0.75

0.7

0.65

0.6

RE
0.55

0.5

0.45

0.4
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
WI (PV)

num sim a* 0.016 0.016 num sim a* 0.008 0.008 num sim a* 0 0
exDP a* 0.016 0.016 exDP a* 0.008 0.008 exDP a* 0 0

Fig. 4—Numerical simulation (“num sim”) and AWC-extended Dykstra-Parsons (exDP) model-prediction results.

–2975 X-axis
–2970
–2980
–2975
–2985
–2980

–2990 –2985

–2995 –2990

–2995
–3000
Z-axis –3000
–3005
–3005

–3010

–3015

–3020

–3025

–3030
0
–200 –175 –150 –125 –100 –75 –50 –25 0
–325 –300 –275 –250 –225
–475 –450 –425 –400 –375 –350 Y-axis

Fig. 5—Water saturation after 4 months, Case “a* 0 0.” Injection is from left to right.

–2975 X-axis
–2970
–2980
–2975
–2985
–2980

–2990 –2985

–2995 –2990

–2995
–3000
Z-axis –3000
–3005
–3005

–3010

–3015

–3020

–3025

–3030
0
–200 –175 –150 –125 –100 –75 –50 –25 0
–325 –300 –275 –250 –225
–475 –450 –425 –400 –375 –350 Y-axis

Fig. 6—Water saturation after 4 months, Case “a* 0.016 0.016.” Injection is from left to right.

August 2018 SPE Reservoir Evaluation & Engineering 711

ID: jaganm Time: 16:08 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 712 Total Pages: 16

The FCC options listed in Table 1 are modeled numerically in a wellbore model, and the (initial) oil inflow-rate distribution along
the production well is plotted on a layer-by-layer basis (Fig. 7).

60

50

Inflow Rate (rcm/d)

40

30

20

10

0
Layer 1 Layer 2 Layer 3 Layer 4 Layer 5
a* 0.016 0.016 a* 0.008 0.008 a* 0 0

Fig. 7—Inflow rate from each layer at the start of production.

Naturally, the stronger the FCC, the more uniform, yet smaller, the inflow rate. It is hard to say which FCC strength is optimal using
this result alone. None of the published methods, such as Prakasa et al. (2015) and Birchenko et al. (2011), that examine the reduced
inflow variation and the associated loss in well productivity predict the actual, long-term recovery or the economic gain from installing
an AWC. By contrast, the extended DP method presented here can do this.
Fig. 4 and Fig. 8 show that the stronger the FCC is, the smaller the volume of injected water is that is required to sweep the reser-
voir. Note the step-like shape of the fractional flow rate in Fig. 8. Each step corresponds to water breakthrough in a layer. The smoother
curve from the numerical simulator is a result of the water-propagation fronts not being exactly piston-like.

1.2

1
Fractional Flow

0.8

0.6

0.4

0.2

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
WI (PV)
a* 0 0 a* 0.008 0.008 a* 0.016 0.016
num sim a* 0 0 num sim a* 0.016 0.016

Fig. 8—Fractional flow rate (i.e., WC at bottomhole conditions) vs. WI for several FCC strength options calculated by the extended
DP method.

It is sometimes useful to quantify how well the FCC reduces flow heterogeneity caused by its adjusting the flow rates, thereby
achieving a more uniform sweep. Lorenz plot is traditionally used for this, where the fraction of the total flow capacity (i.e., flow rate
per net-pay interval) is plotted with respect to either the fraction of the total “static” storage capacity (i.e., PV per interval) or the frac-
tion of the total “dynamic” storage capacity (i.e., streamline’s PV per interval). The net-pay intervals are normally ordered so that the
curve is strictly convex. (Note that, in a layer-cake model, the flow capacity is traditionally calculated as a ratio of the Layer KH to the
Total KH, which is nearly equivalent to the flow rate-based definition and derives from it using Darcy’s law).
The static and dynamic Lorenz plots for the modeled cases are presented in Figs. 9 and 10. Notice how the stronger FCC moves the
curve toward the “ideal”, balanced flow (i.e., flow rate is proportional to the storage capacity) case.
The doubled area between the curve on the Lorenz plot and the “ideal”, unitary-slope straight line represents the measure of hetero-
geneity known as Lorenz coefficient LC. As seen from Fig. 9 and Fig. 10, in the modeled cases, both the static and dynamic Lorenz
coefficients decrease with increasing the FCC strength from 0 to 0.008 bar/(rcm/d)2 to 0.016 bar/(rcm/d)2: The static LC takes the values
of 0.29, 0.23, and 0.22, respectively, whereas the dynamic LC takes the values of 0.21, 0.19, and 0.17, respectively. The reason why the
static Lc shows higher variation than the dynamic Lc is related to the dynamic Lc’s storage capacity in this case, essentially representing
the swept PV (the stream-lines are moving approximately as fast as the layers’ water fronts caused by the nearly piston-like

712 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 713 Total Pages: 16

displacement in one direction) of a layer as well as to the layers’ properties. In the model used (see Table 1), the layers with higher po-
rosity generally have higher permeability (so they have higher flow rates) and lower Sor (so they have higher swept PVs). This is realis-
tic. Thus, the layers’ flow capacity correlates better with their swept PVs (i.e., dynamic Lc is more balanced and shows less variation)
rather than with the total PV (as in the static Lc case).

0.9

Fraction of Total Flow Capacity

0.8

0.7

0.6

0.5

0.4
a* 0 0
0.3
a* 0.008 0.008
0.2
a* 0.016 0.016
0.1 Ideal, balanced inflow

0
0 0.2 0.4 0.6 0.8 1
Fraction of Total Storage Capacity, Static

Fig. 9—“Static” Lorenz plot illustrating how stronger FCC balances the inflow profile better: The corresponding curves tend to
move closer to the “ideal” straight line.

0.9
Fraction of Total Flow Capacity

0.8

0.7

0.6

0.5

0.4
a* 0 0
0.3
a* 0.008 0.008
0.2
a* 0.016 0.016
0.1 Ideal, balanced inflow

0
0 0.2 0.4 0.6 0.8 1
Fraction of Total Storage Capacity, Dynamic

Fig. 10—“Dynamic” Lorenz plot illustrating how stronger FCC balances the inflow profile better: The corresponding curves tend to
move closer to the “ideal” straight line.

The volumetric waterflood forecast shows that it is preferable to select a stronger FCC. However, as expected, the fixed pressure
drop between the wells’ BHPs and a stronger FCC results in a smaller pressure drop within the reservoir and a lower production rate.
The stronger FCC, despite requiring less WI to achieve the target sweep efficiency, may take much longer to inject this WI volume.
Hence, the importance of the AWC analysis being continued into the time domain.
Figs. 11 and 12 show the production forecast. It is now clear that the less restrictive FCC accelerates both oil and water production,
whereas the more-restrictive one constrains them. Option “a* 0.008 0.008” now looks reasonable, giving a relatively fast increase in
RE while significantly delaying water production. The analysis can be continued into, for example, calculating economic parameters
for further quantification of the results, but this is beyond the scope of this work.
Example of AFCD-Completion Modeling. Another interesting exercise is evaluating the potential gain from an AFCD completion.
Note that the value of this completion cannot, by definition, be evaluated with a standalone, wellbore simulator. This is because they
only consider the initial state of reservoir inflow (as explained previously), whereas the economic contribution of an AFCD is its
response to the water breakthrough, which happens some time later. This breakthrough time cannot be reliably predicted without the
use of a dynamic reservoir simulator. The extended DP method, presented here, also can be used reliably, as is shown next.
Figs. 13 and 14 show the performance of an AFCD completion in which the strength doubles from 0.008 to 0.016 bar/(rcm/d)2 after
water breakthrough (“rcm” ¼ cubic meter at reservoir conditions). The figures also show the comparison with the reservoir-simulation
model and the single-strength ICD cases presented earlier.
Note how well the extended DP model prediction matches the numerical simulator. The AICD “a* 0.008 0.016” RE overlays that
for the ICD Case “a* 0.008 0.008”. This is expected: The AICD water response affects only the water-production rate, whereas the oil-
production rate stays the same as for ICD Case “a* 0.008 0.008”. The impact of the increased AICD strength to water is clearly seen in
Fig. 14: The fractional flow-rate (and hence, the WC) curve is between the medium and strong ICD cases.

August 2018 SPE Reservoir Evaluation & Engineering 713

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 714 Total Pages: 16

0.8

0.7

0.6

0.5

0.4

RE
0.3

0.2

0.1

0
0 100 200 300 400 500 600 700
Time (days)
a* 0 0 a* 0.008 0.008 a* 0.016 0.016
num sim a* 0 0 num sim a* 0.008 0.008 num sim a* 0.016 0.016

Fig. 11—RE vs. time for several FCC strength options.

1.2

1
Fractional Flow

0.8

0.6

0.4

0.2

0
0 100 200 300 400 500 600 700
Time (days)
a* 0 0 a* 0.008 0.008 a* 0.016 0.016
num sim a* 0 0 num sim a* 0.008 0.008 num sim a* 0.016 0.016

Fig. 12—Fractional flow vs. time for several FCC strength options.

0.8

0.7

0.6

0.5
RE

0.4

0.3

0.2

0.1

0
0 100 200 300 400 500 600 700
Time (days)
num sim a* 0.008 0.008 num sim a* 0.016 0.016
exDP a* 0.008 0.016 num sim a* 0.008 0.016

Fig. 13—RE vs. time for the AICD completion compared with some ICD completion cases presented earlier.

714 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 715 Total Pages: 16

1.1

0.9

Fractional Flow
0.8

0.7

0.6

0.5

0.4
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
WI (PV)
num sim a* 0.016 0.016 num sim a* 0.008 0.008
num sim a* 0.008 0.016 exDP a* 0.008 0.016

Fig. 14—Fractional flow vs. time for the AICD completion compared with some ICD completion cases presented earlier.

Clearly, the AFCD completion achieves the oil RE as good as its strength to oil, and the water control is as good as its strength to
water. Eltaher et al. (2014) made the same observation. Their recommendation for maximum production efficiency is to design the
AFCCs to be open to oil inflow and very restrictive to water.

Discussion
We have presented the derivation of an extension of the DP method for waterflood analysis in noncommunicating layers to AWCs. The
sandface FCC installed in these wells adds either a nonlinear pressure drop across each layer or the option of zonal closure. An analyti-
cal solution to the introduction of this AWC performance into the DP model was derived, but its solution was performed numerically
because the water-front position could not be expressed explicitly, unlike in the original DP method. The extended DP method can thus
be effectively used for waterflood-performance analysis for wells with arbitrary completion or control.
The important observations, made throughout the paper, are reiterated and explained here.
• It is important to remember that the model ignores gravity and capillary forces, as well as the nonpiston-like displacement in a
given layer. As was shown, this is a reasonable assumption in a heterogeneous reservoir, with the model matching the accurate
numerical-simulation results well (see Fig. 4). The parameters related to the fractional flow rate of water (or WC) in the extended
DP method presented here will have step-like performance. Accurate numerical simulation will predict a smoother change (see
e.g., Fig. 8). Inclusion of the elements of Buckley-Leverett theory within the DP method may be a possible means to address this.
• Eq. 19 cannot be expressed explicitly for x. Hence, an explicit analytical expression for the water front x is not possible for the
extended DP method. This means that several modifications developed by various researchers to the conventional, layer-by-layer
DP method (see Introduction section), cannot be applied to fields developed with AWCs. These include
1. Introducing a statistical description of the reservoir permeability distribution into the solution.
2. Accurate translation of the waterflood-performance prediction into the time domain by integrating the water front x over
the layer. However, our alternative presented here uses the approximate injection rates, proved to be acceptably accurate (e.g.,
Fig. 11).
Osman and Tiab (1981) presented a DP method for composite layers (i.e., layer permeability changes laterally) with conventional
production and injection wells. Their solution has the same quadratic form as the DP model, so it may well be possible to extend our
AWC-DP model to composite layers. Further investigation is required.
• The assumptions made when deriving the waterflood-analysis method in this work include noncommunicating layers, piston-like
displacement within a layer, and others. The applicability of the derived method to the cases where these conditions do not strictly
apply may be the subject of a separate investigation. Such separate, large study can also put quantitative limits on the accuracy/
applicability of the derived method with respect to a range of reservoir and well flow-defining parameters. And as far as the
advanced well-completion design objective is concerned, which is essentially to evaluate the relative benefits of various FCC
types, the applicability limits of the presented method can potentially be wider.
• The solutions presented in this paper assume a constant-pressure difference between the injection and production wells. We have
also explained how to extend this solution to the constant-rate case. Note that the reservoir pressure drop will change markedly
after the breakthrough if the AWC has a strong response to water breakthrough in the production well (e.g., zonal shut-in or
increased FCD strength). This requires that integrals, such as Eq. 23, be split into parts in the constant-rate case.
• This paper provided a list of waterflood-analysis methods for communicating layers. These methods assume a vertical pressure
equilibrium for all layers (i.e., the local pressure derivative at a given distance from the well is the same for every layer). We view
these methods as not being applicable to AWC wells because the AWC purpose is always to create significantly different pressure
drops across each layer.
Our extended DP model matched the accurate numerical-simulation results. It can be used as either a simplified, fast simulator, or
as an analytical model to evaluate the impact of various well-completion and control options on the waterflood development and
efficiency. This model is also a missing link between the various AWC-design methods available today and the AWCs’ long-term-
value evaluation.

August 2018 SPE Reservoir Evaluation & Engineering 715

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 716 Total Pages: 16

Conclusions
The DP method was developed for analysis of the waterflood of a stratified, heterogeneous reservoir with noncommunicating layers.
We have extended the DP method to the situation when the reservoir is developed with advanced wells (i.e., with the wells capable of
controlling flow rate at the zonal level). The extended DP method is fast, simple, transparent, and is particularly useful for predicting,
analyzing, and designing the advanced well performance to maximize oil recovery or field economics. This task is rarely performed in
reality because of the complexity of the existing modeling tools.
We have presented
• Review of the AWC technology.
• Review of the waterflood-analysis models.
• Overview of the original DP method.
• Development of the DP method’s extension to the case of AWC wells. Analytical solutions were presented.
• The extended DP-method application algorithm. It was shown that it can be quickly and easily implemented in, for example, a
spreadsheet or a code.
• Comparison of the extended DP method with the accurate numerical simulation. The extended DP method’s prediction accuracy was
successfully verified.
• Example application of the extended DP method in the AWC completion design and modeling studies. This included the cases with
both fixed and autonomously reacting sandface FCC. The examples illustrated the method’s value and potential.
• Finally, the method’s limitations and potential improvements were discussed. We concluded that the method can be successfully
used on a layer-by-layer basis in noncommunicating layers, but is unlikely to be extended to the situations of communicating layers,
nor to include the permeability-distribution functions.
The waterflood model presented here is a missing link between the various advanced well-completion and control-design methods
and their long-term value evaluation. It can also be used to evaluate the vertical sweep efficiency in the situations when the well’s pro-
duction or injection is controlled at the zonal level, resulting in a more-realistic estimate of the waterflood’s efficiency. The method’s
transparency and simple implementation algorithm can make it a useful tool among well and reservoir engineers.

Nomenclature
(All values are in SI units and at reservoir conditions, unless otherwise stated.)
a ¼ FCC strength defined by Eq. 9
A ¼ effective area perpendicular to flow
b ¼ formation volume factor
B ¼ parameter defined by Eq. 13
C ¼ parameter defined by Eq. 14
fw ¼ fraction flow rate of water (WC at downhole conditions)
F ¼ a constant defined below Eq. 6
h ¼ layer height
k ¼ horizontal permeability
L ¼ distance between wells
M ¼ mobility ratio
P ¼ pressure
q ¼ flow rate
RE ¼ oil recovery efficiency (recovery factor)
S ¼ saturation
Sor ¼ residual oil saturation
Swi ¼ irreducible water saturation
t ¼ time
u ¼ fluid flow velocity
WI ¼ volume of water injected
x ¼ lateral coordinate of the water front
DP ¼ pressure difference
DS ¼ movable saturation (1  Sor  Swi)
k ¼ fluid mobility (i.e., relative permeability/viscosity)
l ¼ viscosity
u ¼ porosity

Subscripts
abt ¼ after breakthrough
bbt ¼ before breakthrough
bt ¼ breakthrough
j, k, R ¼ refers to Layers j, k, or R, respectively
o ¼ oil
or ¼ residual oil (saturation)
r ¼ relative (permeability)
w ¼ water
wi ¼ irreducible water (saturation)

Superscripts
a* ¼ total FCC flow-restriction coefficient for a layer, defined as a* ¼ aproducer þ ainjector
WI* ¼ volume of water injected expressed in reservoir PVs
x* ¼ relative water-front position defined as x* ¼ x/L
k0 ¼ endpoint mobility

716 August 2018 SPE Reservoir Evaluation & Engineering

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 717 Total Pages: 16

Acknowledgments
The authors thank the sponsors of the Value From Advanced Wells Joint-Industry Project at Heriot-Watt University for the support.
The authors also thank Schlumberger Information Solutions for the access to their software.

References
Al-Khelaiwi, F. T., Birchenko, V. M., Konopczynski, M. R. et al. 2008. Advanced Wells: A Comprehensive Approach to the Selection Between Passive
and Active Inflow Control Completions. Presented at the International Petroleum Technology Conference, Kuala Lumpur, 3–5 December. IPTC-
12145-MS. https://doi.org/10.2523/IPTC-12145-MS.
Al-Khelaiwi, F. T. 2013. A Comprehensive Approach to the Design of Advanced Well Completions. PhD thesis. Institute of Petroleum Engineering,
Heriot-Watt University, Edinburgh, Scotland, UK.
Birchenko, V. M., Iu Bejan, A., Usnich, A. V. et al. 2011. Application of Inflow Control Devices to Heterogeneous Reservoirs. Journal of Petroleum Sci-
ence and Engineering 78 (2): 534–541. https://doi.org/10.1016/j.petrol.2011.06.022.
Dake, L. 2001. The Practice of Reservoir Engineering (Revised Edition) (Development in Petroleum Science). Elsevier Science (reprint).
Dyes, A. B., Caudle, B. H., and Erickson, R. A. 1954. Oil Production After Breakthrough as Influenced by Mobility Ratio. J Pet Technol 6 (4): 27–32.
SPE-309-G. https://doi.org/10.2118/309-G.
Dykstra, H. and Parsons, R. 1950. The Prediction of Oil Recovery by Waterflooding. In Secondary Recovery of Oil in the United States. Washington,
DC: American Petroleum Institute.
El-Khatib, N. 1985. The Effect of Crossflow on Waterflooding of Stratified Reservoirs (includes associated papers 14490, 14692, 15043, and 15191).
SPE J. 25 (2): 291–302. SPE-11495-PA. https://doi.org/10.2118/11495-PA.
El-Khatib, N. 1999. Waterflooding Performance of Communicating Stratified Reservoirs With Log-Normal Permeability Distribution. SPE Res Eval &
Eng 2 (6). SPE-59071-PA. https://doi.org/10.2118/59071-PA.
El-Khatib, N. A. F. 2003. Effect of Gravity on Waterflooding Performance of Stratified Reservoirs. Presented at the SPE Middle East Oil Show, Bahrain,
9–12 June. SPE-81465-MS. https://doi.org/10.2118/81465-MS.
El-Khatib, N. A. 2010. Waterflooding Performance in Inclined Communicating Stratified Reservoirs. Presented at the SPE North Africa Technical Con-
ference and Exhibition, Cairo, 14–17 February. SPE-126344-MS. https://doi.org/10.2118/126344-MS.
Eltaher, E. M. K., Sefat, M. H., Muradov, K. et al. 2014. Performance of Autonomous Inflow Control Completion in Heavy Oil Reservoirs. Presented at
the International Petroleum Technology Conference, Kuala Lumpur, 10–12 December. IPTC-17977-MS. https://doi.org/10.2523/IPTC-17977-MS.
Goddin, C. S., Jr., Craig, Jr., F. F., Wilkes, J. O. et al. 1966. A Numerical Study of Waterflood Performance in a Stratified System With Crossflow. J Pet
Technol 18 (6): 765–771. SPE-1223-PA. https://doi.org/10.2118/1223-PA.
Haghighat Sefat, M., Muradov, K. M., Elsheikh, A. H. et al. 2016. Proactive Optimization of Intelligent-Well Production Using Stochastic Gradient-
Based Algorithms. SPE Res Eval & Eng 19 (2): 239–252. SPE-178918-PA. https://doi.org/10.2118/178918-PA.
Henriksen, K. H., Gule, E. I., and Augustine, J. R. 2006. Case Study: The Application of Inflow Control Devices in the Troll Field. Presented at the SPE
Europec/EAGE Annual Conference and Exhibition, Vienna, Austria, 12–15 June. SPE-100308-MS. https://doi.org/10.2118/100308-MS.
Hiatt, W. N. 1958. Injected-Fluid Coverage of Multi-Well Reservoirs With Permeability Stratification. Washington, DC: American Petroleum Institute.
Johnson, C. E., Jr. 1956. Prediction of Oil Recovery by Waterflood—A Simplified Graphical Treatment of the Dykstra-Parsons Method. J Pet Technol
8 (11): 55–56. SPE-733-G. https://doi.org/10.2118/733-G.
MoradiDowlatabad, M., Muradov, K. M., and Davies, D. 2014. Novel Workflow to Optimize Annular Flow Isolation in Advanced Wells. Presented at
the International Petroleum Technology Conference, Kuala Lumpur, 10–12 December. IPTC-17716-MS. https://doi.org/10.2523/IPTC-17716-MS.
Muscat, M. 1950. The Effect of Permeability Stratification in Complete Water-Drive System. Trans., AIME: 349–358.
Osman, M. E. and Tiab, D. 1981. Waterflooding Performance and Pressure Analysis of Heterogeneous Reservoirs. Presented at the SPE Middle East
Technical Conference and Exhibition, Bahrain, 9–12 March. SPE-9656-MS. https://doi.org/10.2118/9656-MS.
Prakasa, B., Muradov, K., and Davies, D. 2015. Rapid Design of an Inflow Control Device Completion in Heterogeneous Clastic Reservoirs Using Type
Curves. Presented at the SPE Offshore Europe Conference and Exhibition, Aberdeen, 8–11 September. SPE-175448-MS. https://doi.org/10.2118/
175448-MS.
Reznik, A. A., Enick, R. M., and Panvelker, S. B. 1984. An Analytical Extension of the Dykstra-Parsons Vertical Stratification Discrete Solution to a
Continuous, Real-Time Basis (includes associated papers 13753,13830, 14873, and 14695). SPE J. 24 (6): 643–655. SPE-12065-PA. https://doi.org/
10.2118/12065-PA.
Snyder, R. W. and Ramey, Jr., H. J. 1967. Application of Buckley-Leverett Displacement Theory to Noncommunicating Layered Systems, J Pet Technol
19 (11): 1500–1506. SPE-1645-PA. https://doi.org/10.2118/1645-PA.
Warren, J. E. 1964. Prediction of Waterflood Behavior in a Stratified System. SPE J. 4 (2): 149–157. SPE-581-PA. https://doi.org/10.2118/581-PA.

SI Metric Conversion Factors

bar Pa 1.000000Eþ05
bar MPa 1.000000E–01
bbl m3 1.589873E–01
B/D m3/d 1.589873E–01
cp Pas 1.000000E–03
darcies m2 9.869233E–13
ft m 3.048000E–01
ft2 m2 9.290304E–02
ft3 m3 2.831685E–02
Pa psi 1.451000E–04
psi MPa 6.894757E–03
psi/ft kPa/m 2.262059Eþ01

August 2018 SPE Reservoir Evaluation & Engineering 717

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002

 REE189977 DOI: 10.2118/189977-PA Date: 30-July-18 Stage: Page: 718 Total Pages: 16

Khafiz Muradov is an assistant professor at the Institute of Petroleum Engineering, Heriot-Watt University. Previously, he worked as
a senior expert at the TNK-BP research institute. Muradov’s research interests include advanced well-completion technology,
downhole production monitoring, well control, and fluid and heat flow in porous media. He has authored more than 20 publica-
tions. Muradov holds BSc and MSc degrees in physics from Novosibirsk State University, and MSc and PhD degrees in petroleum
engineering from Heriot-Watt University. He is an SPE member and has served on various SPE committees. Muradov received the
2012 SPE Ferguson Medal for the best paper published by a young member.
Bona Prakasa is a PhD degree candidate at the Institute of Petroleum Engineering, Heriot-Watt University. Previously, he worked
as a production engineer at Baker Hughes. Prakasa’s research interests include advanced well-completion technology and res-
ervoir-proxy modeling. He has authored several publications. Prakasa holds BSc and MSc degrees in petroleum engineering
from the Bandung Institute of Technology and Heriot-Watt University, respectively.
David Davies is a professor at the Institute of Petroleum Engineering, Heriot-Watt University. Previously, he worked as an engineer
and researcher at Shell. As part of this experience, Davies led theoretical and experimental groups working on technology solu-
tions in virtually all areas of drilling and production operations. His research interests include advanced well-completion technol-
ogy, production and artificial lift, in-well monitoring, and well-control optimization. Davies has authored more than 100
publications. He holds a first degree in chemistry and a PhD degree in theoretical and experimental chemical physics from
Exeter University, UK. Davies is an SPE member and has served on numerous SPE committees; he has also been involved as an
SPE Distinguished Lecturer.

718 August 2018 SPE Reservoir Evaluation & Engineering

View publication stats

ID: jaganm Time: 16:09 I Path: S:/REE#/Vol00000/180002/Comp/APPFile/SA-REE#180002