IPTC-23793-MS

Maximization of the Productivity Index Through Geometrical Optimization of

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

Multi-Lateral Wells in Heterogeneous Reservoir System

Rana S. Amin, Ibrahim M. Abdulwahab, and N. M. Anisur Rahman, Saudi Aramco

Copyright 2024, International Petroleum Technology Conference DOI 10.2523/IPTC-23793-MS

This paper was prepared for presentation at the International Petroleum Technology Conference held in Dhahran, Saudi Arabia, 12 - 14 February 2024.

This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s).
Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s). The
material, as presented, does not necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at
IPTC are subject to publication review by Sponsor Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial
purposes without the written consent of the International Petroleum Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of
not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented.
Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1.214.545.5201.

Abstract
In the recent decades, it is becoming the new norm to drill multi-laterals wells for better reservoir coverage,
improved productivity index, and superior financial return on investment. Optimum design and placement
of this type of wells can be studied through rigorous modeling. This study presents a multi-parametric
optimization approach that optimizes the design of multi-lateral wells and maximizes the contact with highly
productive hydrocarbon zones in the reservoir.
Advanced transient numerical models built in 3D by incorporating the dynamic data into the geological
model to mimic the transient-pressure behaviors of multi-lateral wells for a given set of geometrical
parameters. Such compliant models with dynamic data indeed capture the reservoir description and
dynamics. The optimization process is subject to a number of input variables, such as maximum number of
laterals, minimum spacing between wells, and maximum lateral length based on the reservoir characteristics.
The multi-parametric optimization generates multiple realizations with different patterns. The productivity
index of each pattern is calculated to look for the best multi-lateral well to be drilled.
This study presents a numerical methodology of geometrically optimizing multi-lateral wells. Several
local optimizations are performed around each main wellbore to place the lateral sections, and to determine
the number of lateral sections. Main objectives of all these realizations are to minimize the competition
among the lateral sections, and maximize the drainage area, which do subsequently affect the well
productivity index. In permeable reservoirs, interference or competition among the lateral sections comes
very quickly, and however, in tight reservoirs, such an occurrence is delayed. After running each realization,
well productivity index is calculated, and the productivity index graph is generated against each constraint
based on the advanced transient numerical model. Graphical presentation of the productivity index helps
decide on the best optimum multi-lateral design. Additional sensitivity analysis is presented to show the
impact of reservoir heterogeneity, lateral-section lengths, lateral spacing, number of lateral sections, and
the impact of offset producers or injectors.
This workflow will help design the most optimum multi-lateral well with a maximum productivity index
under different reservoir conditions in an actual dynamic environment. The proposed workflow has been
tested successfully.
2 IPTC-23793-MS

Introduction
Multi-lateral wells have been widely applied as a superior production technology to access the difficult
geological formations and unconventional reservoirs. Multi-branched horizontal wells in producing oil
have numerous advantages compared to conventional horizontal wells, such as fast recovery and enhanced
sweep efficiency, especially in encountering challenges of plausible water or gas conning. Moreover, larger
drainage area and higher single-well production rate can be obtained using fishbone wells. Xing et al. (2012)

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

have reported that the fishbone-shaped, multi-lateral wells have performed with much better productivity
as much as a 10-fold increase in production rate than the multi-fractured horizontal wells in relatively low
permeable reservoirs.
Generally, a multi-lateral well is defined as a well with multiple branches in the lower bore-hole targeting
the pay zones in the same layer or different layers. Multi-lateral wells can be either complex, multi-branched
wells or simple, dual-opposing lateral sections. According to Bosworth et al. (1998), with the main wellbore
being vertical, multi-lateral wells can be classified into some common types as illustrated in Fig. 1. They
consider a multi-lateral well to be a successful one, when it replaces several vertical wellbores to reduce
overall drilling and completion costs, increase production, and provide more efficient drainage of the
reservoir.

Figure 1—Classification of multi-lateral wells (modified after Bosworth et al. 1998).

Although the basic forms of multi-lateral wells have been used since the nineteen-fifties, these wells
were not so popular back then because the early drilling and completion methods did not favor their use in
many well applications. With improvements in drilling methods, drilling tools, well construction techniques
and completion, more emphasis is currently being placed on the use of multi-lateral well applications in
petroleum production, especially in the production of heavy oil reservoirs, fractured reservoirs, matured
field development, faulted and compartmentalized reservoirs, laminated formations, tight reservoirs, and
thin layered formations. Since more and more multi-lateral wells are currently being utilized, and some
fields are being developed exclusively with multi-lateral technology, the multi-lateral well architecture,
completion selection and factors affecting the use of this technology needs critical analysis. Delamaide
(2021) has listed advantages of multi-lateral wells over the corresponding counterparts of horizontal wells,
but also acknowledged the challenge of pressure interference among lateral sections. Numerical modeling
IPTC-23793-MS 3

for geometrically optimizing multi-lateral wells will further help design elegantly the multi-lateral sections,
and get the full benefit out of it.
In this paper we are interested in multi-branched, forked multi-lateral and stacked lateral sections. Design
of these lateral sections based on the reservoir parameters is a complex process. In order to achieve the
optimum design, multiple parameters and scenarios need to be considered so that economic benefits of
multi-lateral wells are fully realized. Bosworth et al. (1998) acknowledged the benefits of multi-lateral wells

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

in terms of increased production, increased reserves, and overall reduction in reservoir development costs.
In our study, we have accounted for the reservoir characteristics in addition to the other numerous supporting
parameters in analyzing the performance of multi-lateral wells.

Methodology
Currently in many fields, development economics favor the use of multi-lateral wells where a minimum
number of wells, each having a maximum contact with the reservoir, are drilled for oil or gas production.
Drilling several branches into a reservoir from the main wellbore offers potential benefits in terms of
improving drainage, productivity and well economics. Multi-lateral techniques are becoming popular for
both onshore and offshore operations, and can now be seen as one of the most important innovations in the
industry for oil and gas field development. Numerical modeling for geometrically optimizing multi-lateral
wells will further help design the multi-lateral sections smartly, and get the full benefit out of it. Lux et al.
(2016) have demonstrated a procedure for rigorous modeling of multi-lateral wells in flow simulators. Al-
Qahtani and Berlow (2021) have linked the identification of sweet spots in the reservoir with the placement
of an optimum multi-lateral well. On a similar note, Al-Ismael et al. (2022) have utilized a mathematical
optimization formulation to seek for maximization of the reservoir contacts with multi-lateral wells.
As the primary objective of drilling multi-lateral wells is to maximize the productivity or the injectivity,
any additional investment on multi-lateral wells must be justified technically. Therefore, proper geometrical
optimization of multi-lateral sections is required. With the recent technology, drilling of multi-lateral
sections is not a difficult job. But placing a perfect multi-lateral section to achieve a maximum productivity
out of it is a challenge. As such, a computer-aided, reservoir modeling workflow for geometrical
optimization of multi-lateral wells is utilized. Several local optimizations are performed around each main
wellbore to place the lateral sections, and to determine the number of lateral sections to meet the production
targets. Main objectives of investigating all these realizations are to minimize the competition among the
lateral sections, and maximize the drainage area, which subsequently do contribute to the well productivity
index.
In permeable reservoirs, pressure interference or competition among the lateral sections over the common
drainage area is sensed very quickly. However, in tight reservoirs, such an occurrence is delayed. After
running the reservoir model on each realization, the well productivity index is calculated. A productivity
index graph is then constructed from the results of the advanced transient reservoir models for each category
of constraints. None of the earlier studies reported in the literature have focused on using the productivity
index of a multi-lateral well as a criterion for judging the best option for drilling this kind of wells. To select
the best option for a multi-lateral design intuitively, we have presented productivity indices graphically.
Additional sensitivity analysis is presented to show the impact of reservoir heterogeneity, lateral-section
lengths, lateral spacing, number of lateral sections, and the impact of offset producers and injectors.
The methodology to geometrical optimize the multi-lateral well is applied in following five steps:

• Selection of well location to meet production target

• Optimum lateral spacing scenario

• Optimum lateral configuration

4 IPTC-23793-MS

• Optimum design lateral length

• Injector to producer response time based on selected location and distance

Main objective behind this optimization is to design and place the well so that maximum productivity or
injectivity index could be achieved. The productivity index (PI) is defined as:

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

(1)

where
q = Production rate, STB/d,
= Average reservoir pressure, psia,
pwf = Final bottom-hole flowing pressure, psia.

Workflow
To apply the geometrical optimization of multi-lateral wells, a systematic workflow has been adopted. This
workflow involves six major steps to accomplish the complete task as presented in Fig. 2.

Figure 2—Multi-lateral geometrical optimization workflow.

Each of the six major steps in the workflow is presented below.

Geological/Numerical Sector Model Preparation

To optimize the multi-lateral well, geological model is the key requirement in order to capture the reservoir
heterogeneity, and its impact on the well productivity. In case a geological model is not available, a numerical
model can be built by utilizing the offset well logs and dynamic data. Such a numerical model may not
be as effective as the suggested geological model, but it will be useful in capturing some aspects of the
reservoir behavior.

Transfer to Numerical Simulator

A geological sector model is transferred to a computer-aided, numerical simulator, which can simulate the
pressure behaviors due to the production through the subject multi-lateral well. The simulated pressure
behavior in each scenario leads to computing the corresponding productivity index of the well, a parameter
IPTC-23793-MS 5

considered in optimizing the multi-lateral sections geometrically. In this study, we have utilized an advanced
transient modeling software to determine the best options. This numerical package provides a platform to
compare different scenarios, and helps produce consistent results as the model output.

Data Incorporation (Reservoir, Well, Production)

All the reservoir (PVT, SCAL, fluid contacts, etc.), well completion details, and expected production rates

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

are incorporated in order to simulate all the operating conditions in line with original field dynamic model.

Initialization
After incorporating all the required data in the geological model, the model is initialized. From this point
on, the reservoir model should be ready to simulate all the plausible scenarios for comparison.

Sensitivity Analysis
The sensitivity analysis based on the initial results is run to optimize the geometrical configuration of the
multi-lateral well. It will help identify the impacts of different parameters of interest, and get the best multi-
lateral design with a maximum productivity index.

Final Results
The results of all sensitivity analyses are compared, and the best geometrical configuration with a maximum
productivity index is selected as a clear front runner in the comparison.

Multi-lateral Geometrical Optimization

This is the most important aspect in this study. To complete the multi-lateral geometrical optimization, there
are four progressive phases in succession, each of which is inter-dependent and contributes to the decision
making. These phases are listed below.

• Selection of well location

• Identification of optimal lateral spacing and configuration

• Design length optimization

• Injector to producer spacing

Each of the phase is discussed briefly in the following sub-sections.

Selection of Well Location

Before going for the multi-lateral optimization, it is important to find if the selected well location can meet
the target production rate. Sometimes well locations are selected in advance but the field dynamics may
have changed in the meantime, which may impact the production goals. In this study, the future multi-lateral
well is placed at the proposed location in the model, and simulate the target production rates as illustrated
in Fig. 4. If the reservoir model suggests that the well would produce a lot of water, as presented in Fig. 5,
the proposed location cannot be accepted for drilling the well. So, a new, alternative well location needs to
determined. Sensitivities due to multiple locations in the vicinity of the proposed location are examined to
select the best location that can meet the production objectives. This idea is demonstrated in Fig. 4.
6 IPTC-23793-MS

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

Figure 3—Multi-lateral well utilized for geometrical optimization.

Figure 4—3D and 2D views of the model, proposed location and the vicinity where production sensitivities are performed.

Figure 5—Proposed well location is producing more water as compared to oil.

After running the sensitivity due to multiple well locations to achieve the target production rate, the
Well-3 location in Fig. 4 is selected as the optimum location. Having met the production targets in the new
location, the well demonstrates the updated performance as shown in Fig. 6.
IPTC-23793-MS 7

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

Figure 6—Well-3 location is found to be the most optimistic, meeting production target.

Identify Optimal Lateral Spacing and Configuration

Identification of optimal spacing in between the lateral sections is important in order to minimize the
competition among the lateral sections over the common drainage area. This drainage area competition
does eventually impact the productivity index (PI). The objectives of drilling multi-lateral wells are to
maximize the reservoir contact and the PI. Simulation scenarios are generated with different configurations
and spacings for tri- and dual-lateral wells by yielding to the given reservoir heterogeneity. The PI is
calculated for each simulated scenario as shown in Figs. 7 and 8. A graphical presentation with respect to
the PI helps the reservoir analyst select the optimum multi-lateral well to be drilled.

Figure 7—Potential configurations of plausible dual- and tri-lateral wells.

Figure 8—PI calculation considering tri- and dual-lateral scenarios with different spacings.
8 IPTC-23793-MS

To identify the most optimal configuration of the multi-lateral well, the PI is calculated for different
configurations, considering dual- and tri-lateral scenarios.
Lateral configuration is primarily driven by the reservoir heterogeneity. In some cases, based on the
reservoir quality, a dual-lateral well may demonstrate a better PI, and consequently helps save a lot of
resources, including manpower and investment. Each of these options is discussed below.
Tri-lateral. Refer to Fig. 8. The PI in the tri-lateral option increases by about 60% due to increasing

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

the spacing from the 100-m spacing to the 400-m spacing. For increasing the spacing from the 400-m
spacing to the 600-m spacing, the increase in PI is only about 10%. The tri-lateral scenario shows the lowest
productivity index at the 100-m spacing due to the intense pressure interference among the three lateral
sections. As the two other lateral sections, L2 and L3, are moved away from the main wellbore L1, the
pressure interference among the lateral sections eases, and the productivity index increases.
Dual-lateral. Refer to Fig. 8. In the effort for optimization of the multi-lateral well configuration, it is
observed that the dual-lateral well with lateral sections L1 and L2 adds about 100% to the PI value due to
increasing the spacing from the 100-m spacing to the 600-m spacing. Yet, the total PI is less than that of the
tri-lateral well. The dual-lateral option due to the lateral sections L1 and L3 does not contribute positively
to the PI, and this value continues to remain at a maximum value as the spacing increases. This behavior
suggests that the reservoir heterogeneity plays a key role here, and the reservoir quality degrades towards
the lateral section L3. Any increase in the spacing, between the lateral sections, is supposed to impact the
overall performance of well PI positively. But increasing the spacing in the dual-lateral option with the
lateral sections L1 and L3 to a value above the 400-m spacing, there is no additive PI gained as the lateral
section L3 falls in the lower-quality reservoir area. In the dual-lateral option with lateral sections L1 and
L2, it appears that increasing the spacing to more than the 400-m spacing between the lateral sections will
bring more improvement to the PI value. But in the case of the dual-lateral option with the lateral sections
L1 and L3, there is no added benefit if the spacing is increased to a value more than the 400-m spacing.
Comparison and selection. The above process has clearly identified the well configuration that can give
the best PI value. This will help optimize the multi-lateral configuration by considering other reservoir
and operational parameters. In the presented example, the 400-m lateral spacing is the most optimum one
because any further increase in spacing results in a very marginal or no increase in the PI. Although any
further increase in the lateral spacing may contribute to the PI values marginally, there will be additive
reservoir drainage area to be covered by each lateral. As such, fewer wells can be drilled in reservoir.
However, the reservoir sweep efficiency may be compromised.
After selecting the optimal spacing between the lateral sections, the next important step is to identify the
best design lengths for all the lateral sections to achieve a maximum PI.

Design Length Optimization

After identifying the optimal configuration and lateral spacing based on the PI, the next important phase is to
look for the design length to further maximize the PI. Drilled lengths play a major role in well productivity.
In a good-quality, homogeneous reservoir, the effective well lengths are high, which eventually boosts the
PI values of the completed wells. In highly heterogeneous reservoir systems, an increase in the drilled length
may not always be impactful. Bosworth et al. (1998) have acknowledged that the reservoir properties drive
the placement of multi-lateral sections in the reservoir. In other words, an existing drilled length may not
have a linear relationship with the PI value. If the reservoir quality degrades, an increase in design length
may indeed reduce the PI. The situation is reversed in the good-quality reservoirs. Therefore, sensitivity
analysis for the design length is required to find the best option of the wells lengths to be drilled. One
other important factor needs to be considered while running the sensitivity scenarios for design length
is the reservoir quality. In a better-quality reservoir, interference between the lateral sections can happen
faster compared to that in the lower-quality reservoir. This interference between the lateral sections may
IPTC-23793-MS 9

cause the competition, which consequently reduces the PI of the well. It is not obvious that any increase in
design length will help improve the PI value. Sometimes due to reservoir heterogeneity, it is possible that
an increase in the design length may not cause any further improvement in the PI values.
Fig. 9 presents the sensitivity analysis performed for different design well lengths, where in the design
lengths for all lateral sections are considered equal for the presented cases. Fig. 9 suggests that for a tri-
lateral well with the 7,000 ft of design length in each lateral section is the most optimum. In the dual-

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

lateral option, the 10,000 ft of design length for each lateral section will also give the same PI. But due
to operational limitations and future maintenance of the well, this option with such a long well length
might pose a challenge in the long run. Therefore, considering the calculated PI values based on the model
responses, the most optimum design length scenario is chosen to be the one with 7,000 ft of each lateral
length, and a tri-lateral well is the most optimal geometrical configuration.

Figure 9—PI calculation considering the tri- and dual-lateral scenarios with different design lengths.

Injector to Producer Spacing

In some reservoirs, enhanced oil recovery methods are employed by injecting different fluids into the
reservoirs to maintain the reservoir pressure and improve the oil recovery. Main idea of these recovery
methods is to provide pressure support to the producers, displace the hydrocarbons, and sweep them towards
the producers. Looking at the objectives of these injection wells, it is vital to know that how these injection
wells will impact the producers. The spacing between the injection and the producer becomes the key driver.
This effort is supported by estimating the pressure responses at the nearby producers for a given reservoir
quality. Based on such pressure responses, the locations of injection and production wells are planned and
drilled at.
In geometrical optimization of the multi-lateral wells, the injection pressure response, and the degree
of pressure support are analyzed to maximize the PI, and achieve a maximum recovery target. Multiple
sensitivity scenarios have been performed, considering the injector to producer spacing, and the injection
rates. Pressure response for each spacing and injection rate is predicted from the model, and presented
below in Fig. 10. Other important factors that need to be considered while designing the producer to injector
spacing are the water production rate and the water cut at the producer. The water production at the producer
should be monitored, and the spacing between injector and producer needs to be optimized in such a way
that the water cut at the producer remains within the design limit. A well-built transient model can facilitate
this key reservoir engineering tool.
10 IPTC-23793-MS

Downloaded from http://onepetro.org/IPTCONF/proceedings-pdf/24IPTC/2-24IPTC/D021S063R006/3366402/iptc-23793-ms.pdf by Universidad Nacional De Colombia user on 20 March 2024

Figure 10—Spacing scenarios optimization between multi-lateral and an injection well.

Conclusion
The increasing use of multi-lateral wells calls for more sophisticated techniques to optimize the well for a
smart and efficient field development strategy. The multi-lateral well placement is an essential technology
to achieve the maximum reservoir contact and optimal recovery of hydrocarbons. This new workflow
has been introduced to geometrically optimize the multi-lateral wells, which requires the minimization
of interference, and the reduction in competition over the drainage area among the lateral sections.
Optimization has been done by evaluating the effects of the number of lateral sections, spacing among lateral
sections, and design lengths for fork-type, multi-lateral wells on the PI. Achieving a maximum PI from a
geometrically-optimized well is cost effective, and will improve the recovery of hydrocarbons. The last part
of this study deals with the producer to injector spacing so that the reservoir optimally gains the external
recharge with the fluid injection. Current computer-aided, reservoir modeling solutions can be utilized in
designing an optimal multi-lateral well geometry.

SI Metric Conversion Factors

bbl × 1.589 873E–01 = m3

ft3 × 2.831 685E–02 = m3
psia × 6.894 757E+00 = kPaa

References
Al-Ismael, M.A., Al-Turki, A.A., and Al-Saffar, A.H. 2022. Multi-Parametric Optimization of Multilateral Wells for
Optimum Reservoir Contact. Presented at the SPE EuropEC - Europe Energy Conference Featured at the 83rd EAGE
Annual Conference and Exhibition, Madrid, Spain, 6–9 June. SPE-209647-MS. https://doi.org/10.2118/209647-MS.
Al-Qahtani, G.D., and Berlow, N. 2021. Large Scale Placement for Multilateral Wells Using Network Optimization.
Presented at the SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, 28 November–1 December.
SPE-204803-MS. https://doi.org/10.2118/204803-MS.
Bosworth, S., El-Sayed, H.S., Ismail, G. et al. 1998. Key Issues in Multilateral Technology. Oilfield Review 10 (4): 14–28.
Delamaide, E. 2021. Application of Multilateral Wells for Production and Enhanced Oil Recovery: Case Studies from
Canada. 2021. SPE Res Eval & Eng 24 (4): 952–974 SPE-206745-PA. https://doi.org/10.2118/206745-PA.
Lux, M., Szanyi, J. and Tóth, T.M. 2016. Evaluation and Optimization of Multi-lateral Wells Using MODFLOW
Unstructured Grids. Open Geosciences 8 (1): 39–44. 3 February. https://doi.org/10.1515/geo-2016-0004.
Xing, G., Guo, F., Song, C. et al. 2012. Fishbone Well Drilling and Completion Technology in Ultra-thin Reservoir, IADC/
SPE Asia Pacific Drilling Technology Conference and Exhibition, Tianjin, China, 9–11 July. SPE-155958-MS. https://
doi.org/10.2118/155958-MS.