Segregating Laterals for Efficient Gas Re-Injection in Shale Plays Using Smart Completion

Abstract

There is a strong global demand for oil and gas resources, and forecasts indicate robust growth in oil demand in the coming years. Meeting this demand necessitates the exploitation of unconventional resources and enhancing the recovery of existing oil and gas fields. Field trials indicate that traditional gas injection in shale wells has low sweeping efficiency. Emerging technologies play an exceedingly significant role in solving the challenges of gas EOR in shale and tight formations. Among these advancements, smart or intelligent well technology has emerged as a promising solution to enhance field development outcomes. This study focuses on improving gas flooding efficiency in the Bakken formation by utilizing smart completions to reduce the gas–oil ratio (GOR) and increase oil recovery. An economic assessment of gas re-injection is conducted, considering both gas storage and enhanced oil recovery, with analysis incorporating capital expenditures, operating costs, and revenue from increased production. Reservoir simulations were employed to determine the most effective gas injection scenarios for maximizing recovery and storage efficiency. Simulation results demonstrate that 20% of perforated laterals account for 80% of the injected gas. To address this challenge, this work proposes using smart completions to segregate lateral sections, thereby optimizing gas injection efficiency, and unlocking additional oil in tight formations. Segregating horizontal laterals for gas re-injection using smart completion technology can achieve gas injection efficiencies of up to 0.25 barrels per Mcf of gas injected, with lower incremental gas production. The optimal injection rate is between 1 MMcfd and 3 MMcfd, with an injection period ranging from one to three years. It was also found that injecting gas into the toe section results in high bottom hole pressure but lower oil recovery due to reduced gas injection efficiency. From an economic perspective, the project yielded favorable outcomes, with a positive net present value (NPV) at a 7% discount rate. Even at lower oil prices (USD 50 per barrel), the Internal Rate of Return (IRR) was calculated to be 170%, indicating strong potential profitability.

1. Introduction

The global demand for oil and gas resources is experiencing significant growth due to factors such as population increase, industrial activities, and improved living standards [1]. Shale reserves and tight formations have become crucial contributors to oil and gas production in the United States [2]. Currently, approximately 65% of U.S. production comes from these unconventional sources, and this percentage is projected to reach 75% by 2050 (EIA). North America accounts for 23% of recoverable unconventional resources [3]. Although horizontal drilling and multistage hydraulic fracturing techniques have been successfully applied in shale reservoirs and tight formations [4,5,6,7], the recovery factor remains relatively low. Primary depletion of oil in these reservoirs typically results in a recovery factor of less than 10% [8], leaving a significant amount of hydrocarbons unrecovered. One common method employed in the development of unconventional reservoirs is the huff-n-puff technique using natural gas, CO₂, and N₂ [9,10,11]. Field trials of natural gas huff-n-puff applications in the Eagle Ford formation, as analyzed by Hoffman [12], demonstrated the technique’s viability in extending field life and increasing oil recovery. Similarly, Pospisil et al. [13] evaluated a rich gas multi-well cyclic huff-n-puff pilot in the Bakken formation, finding that a three-month injection cycle with a gas injection rate of 3 MMcfd could potentially yield an incremental recovery ranging from 4% to 26%. Additionally, Sanaei et al. [14] investigated the application of CO₂ huff-n-puff in the Bakken and Three Forks reservoirs, exploring its effectiveness in enhancing oil recovery. The introduction of solvent into the fracture system enhances the recovery rate by elevating the diffusion coefficient and effective density gradient in the reservoir. Notably, the solvent is not directly injected into the matrix but rather diffuses through a counter diffusive process, with oil being produced into the wellbore as the solvent diffuses into the matrix [15]. Overall, these studies and field trials demonstrate the ongoing efforts to apply huff-n-puff techniques with various gases to maximize oil recovery from unconventional reservoirs. It was substantiated that number of cycles, gas type, and gas injection rate are the fundamental parameters to maximize incremental oil recovery. Co-injection of produced gas and water-surfactant mixture is an innovative huff-n-puff approach. A pilot project was conducted with main objectives to re-pressure the reservoir above the MMP, prove the concept of using water co-injection to build hydrostatic pressure to inject gas at low surface pressures and to improve gas conformance, and use a surfactant to enhance oil recovery through rock wettability alteration [13].

Field trials indicate that traditional gas injection in shale wells has low sweeping efficiency [16]. Emerging technologies play a paramount role in the oil and gas industry [17,18]. Among these advancements, smart or intelligent well technology has emerged as a promising solution to enhance field development outcomes. This technology has been specifically developed to facilitate improvements in the overall field development process. Most of the time, smart completion is associated with high CAPEX. On the contrary, it allows for accelerated cash flow, reducing CAPEX and OPEX in many ways. For instance, availability and accessibility of real-time data monitoring achieved via the pressure and temperature permanent downhole gauges is crucial to ensure that the changes in well and reservoir, during flowing or shut-in condition, can be captured for various analysis. It can allow engineers to evaluate the well lifting performance and propose solutions such as improving artificial lift efficiency to gain more barrels [19]. Smart completion was successfully installed in a water injector well. The primary objectives were centered around enhancing reservoir management, reducing complexity in well construction, and achieving cost optimization [20]. In the past decade, numerous pilot projects have been implemented in the Bakken region to evaluate various EOR technics and strategies. The reported results of these projects were analyzed in [21]. Overall, the findings indicated early breakthrough and low sweeping efficiency. One notable pilot project involved gas flooding, although it was discontinued after just 55 days. However, this pilot project demonstrated the feasibility of using rich natural gas as the injection fluid, which can be more readily available compared to CO₂ in unconventional oil plays. Additionally, the cost of natural gas in the Bakken is currently lower than that of CO₂. Another advantage of using natural gas is its ability to address the issue of flaring, which remains a concern in certain parts of the Bakken region.

In another study [22], authors addressed the technical challenges and limited success of huff-and-puff and continuous injection techniques in tight oil formations for fundamental reasons. Huff-and-puff methods encounter difficulty due to the need for substantial fluid transfer into and out of the tight oil formation during each injection-soaking-production cycle, which becomes challenging because of the low permeability of the formation. Similarly, continuous injection using fractured horizontal wells for injection and production faces the issue of short circuits caused by fractures that communicate directly from injector to producer. Aiming to tackle these challenges, this study proposes the utilization of a single well with a dual-conduit completion. This completion design enables continuous injection and production by spatially alternating injection and production fracture sections, which are isolated from each other by packers. However, such a technique requires a sophisticated completion with multiple remotely activated valves that enable asynchronous injection and production from the spatially alternating fracture sections within a single well.

This paper proposes the implementation of smart completion in horizontal wells drilled in the Bakken to segregate the gas injection along the lateral section. The objectives are to improve the gas injection efficiency, reduce the produced GOR and increase oil recovery. Thorough reservoir simulation is undertaken to identify the most favorable gas injection scenario. This involves evaluating CAPEX, OPEX, and potential revenue from oil recovery. By considering these factors, an assessment is made to evaluate the most promising approach for gas injection.

2. Project Overview

Currently, the state of North Dakota, U.S., produces over 1.1 million bbl/d of crude oil. However, this amount is only a small fraction of the estimated total oil in place of approximately 600 billion bbl [23]. The sharp production decline rate of Bakken and Three Forks horizontal wells (85% over the first 3 years) results in a low primary recovery factor that is estimated to be less than 10% of original oil in place (OOIP). Therefore, each 1% increase and/or long-term stable production of oil in Bakken and Three Forks wells would yield over a billion barrels of additional oil [13]. The combination of these factors makes these formations critical targets to improving oil recovery.

2.1. Intelligent Completion

The completion design integrates permanent downhole sensors and surface-controlled downhole flow control valves, enabling operators to continuously monitor, evaluate, and actively manage production or injection in real time without the need for well interventions. Data collected from the downhole sensors is transmitted to the surface for local or remote monitoring purposes. This study uses intelligent completion to segregate the horizontal lateral for gas injection. Intelligent completions are employed to optimize injections and achieve improved sweep by controlling flow into specific zones. The primary objective is to enhance the distribution of injected fluids for more efficient sweeping of reservoir.

2.2. Concept

The study area is situated in Williams County and consists of four leases developed using three multi-well pads (refer to Figure 1). Each well features a lateral length of 10,000 ft. The selected leases include Borsheim Trust, Orville, Cherrey, and Charles. One of the multi-well pads comprises four wells: Borsheim Trust #1H, Borsheim Trust #2H, Orville #1H, and Orville #2H. Across all four leases, the wells produce an average of 30–40 barrels of oil per day (bpd), with a gas oil ratio (GOR) of approximately 1.1 Mcf/bbl.

Borsheim Trust 2H was selected for a continues gas injection scheme. Three gas volumes were modeled to determine the effect of different gas injection rates.

• Gas injection of 1 MMcfd.
• Gas injection of 2.8 MMcfd, total production of gas gathering branch.
• Gas injection of 13 MMcfd, maximum gas that can be injected into a well.

Three injection periods:

• 3 months then the gas injector resumes production.
• 1 year then the gas injector resumes production.

The intelligent completion allowed the segregation of the lateral for gas injection. The gas was injected in sequence starting from the toe section of the lateral (Figure 2). Three cases were considered:

• Gas injection through normal completion.
• Gas injection through intelligent completion. The lateral is segregated into two sections, toe and heel.
• Gas injection through intelligent completion. The lateral is segregated into three sections, toe, middle and heel.

The injection volume is anticipated to impact incremental oil recovery, and, in accordance with regulatory constraints, it is critical to assess its effects on neighboring wells.

3. Reservoir Model Construction and Injectivity Constraints

In this simulation study, Petrel and Eclipse 2024.2 software were used. The Equation of State and refinement grids were selected to simulate the compositional interaction of the reservoir fluid with the injected gas. Reservoir fluid composition and gas injection composition were taken from [24].

In this study, it was assumed that each well would have ten hydraulic fracturing (HF) stages, with a spacing of 1000 feet between each stage. The relative permeability curves used in the analysis were adapted from [25]. The initial water saturation was assumed to be equal to the residual water saturation, which was set at 0.2. The pore compressibility of a consolidated limestone reservoir was estimated using the Newman equations. Stress-dependent permeability was calculated using the equation proposed in [26], specifically Equation (1). The values of Ko and γ are consecutively, 0.0083 md and 3 × 10⁻⁴, respectively, based on the findings in [27].

$$K=Ko\times \mathrm{e}\mathrm{x}\mathrm{p} \left[\right(-\mathsf{\gamma }\left(\mathrm{P}\mathrm{e}\mathrm{f}\mathrm{f}-\mathrm{P}\mathrm{o}\right)]$$

(1)

K denotes permeability under net confining pressure P_eff. K_o is permeability under atmospheric pressure (14.5 psi). γ is the pressure sensitivity coefficient.

Initial reservoir pressure and average petrophysical characteristics are summarized in

Table 1. Reservoir properties.

Reservoir Specific Information	Unit
Reservoir Top (ft)	10,000
Average Thickness (ft)	55
Average Porosity	6
Average Permeability (md)	0.009
Initial Pressure (psi)	7000
Water Saturation (%)	50
Temperature (° F)	240
Oil Gravity (° API)	45

. Laboratory experiments demonstrate that produced gas acts as a highly effective solvent, particularly for Middle Bakken samples [28]. Rich gas mixtures consisting of 70% methane, 20% ethane, and 10% propane from the Bakken formation can reach minimum miscibility pressure (MMP) at relatively low pressures, approximately 2450 psi. Using PVTP 13.0 software, the MMP for the produced gas was estimated to be 2335 psi.

History matching results showed a strong agreement between the simulated and observed data. Figure 3 presents the cumulative oil, water, and gas production, as well as the bottom hole pressure for Orville 2H. During the natural depletion phase, all wells initially exhibited high oil production rates, followed by a sharp decline a few months after production commenced. This trend is characteristic of horizontal, multistage hydraulically fractured wells drilled in tight formations.

Ultra tight formation such as the case of the Bakken is characterized with very low permeability, which makes the injectivity of the formation challenging. Several gas injection cases were conducted to guide the conceptualization of the EOR schemes.

Figure 4 illustrates streamline results of continues gas injection of 10 MMcfd in Borsheim Trust 2H through normal completion. As can be seen, the injected gas has flooded the center of the multi-pad where the deferential pressure is the highest. As a consequence, lower swiping efficiency and rapid gas breakthrough to the offset wells. The main objective of this study is to increase the gas injection efficiency by deploying smart completion to control the injection intervals.

Below is a summary of the results obtained for varying the length and location of the injection interval:

• Injection in the last 10% of the lateral (1000 ft near the toe): max gas injection was 150 Mcfd. BHP reached 12,000 psi.
• Injection in the last 20% of the lateral (2000 ft near the toe): max gas injection was 600 Mcfd. BHP reached 12,000 psi.
• Injection in the last 30% of the lateral (3000 ft near the toe): max gas injection was 2.5 MMcfd. BHP reached 12,000 psi
• Injecting in the middle 10% of the lateral: max gas injection was 1 MMcfd. BHP reached 12,000 psi.
• Injecting in the middle 20% of the lateral: max gas injection was 2.2 MMcfd. BHP reached 12,000 psi.
• Injecting in the first 20% of the lateral (near the heel): max gas injection was 10 MMcfd. BHP reached 8000 psi.

To summarize,

• Injecting the gas in the toe section of the lateral can be technically challenging. Based on the simulation results, injecting the gas in the last 30% of lateral will not increase the production, and it can be considered as waste of energy. Therefore, in all upcoming injection cases, we excluded the last 30% of the horizontal section.
• At least one-third of the lateral section must be kept open to be able to inject the amount of gas, 3 MMcfd or more. Therefore, in all upcoming injection cases, the lateral was segregated to no more than three sections.
• The heel section is responsible for almost 80% of the injected gas through normal completion.

4. Numerical Simulation Results

• Continuous injection of 1 MMcfd for three months:

The first set of gas injection schemes was designed to inject 1 MMcfd over a three-month period. Two cases were considered (Figure 5), black is the case with no gas injection and red is the injection through normal completion. Green is the injection through smart completion in which the lateral was segregated into three sections, starting the injection in the toe section, then the middle section and finally the heel section. As can be seen, both injection cases gave approximately the same incremental oil and slightly less gas in the case of injection through smart completion. Higher oil and gas rates were obtained as soon as the injector was switched back to production.

Table 2. Simulation results of 1 MMcfd injection (3-month injection period).

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)
1 MMcfd Nor_Complet	59%	16	0.18	1.1
1 MMcfd Smart_Complet 3 Sections	59%	15	0.17	1.1

presents the simulation results, showcasing various parameters and their respective values. One of the parameters used to evaluate the effectiveness of the injected gas is the EOR (enhanced oil recovery) efficiency. This efficiency is calculated by determining the incremental oil achieved for each Mcf (thousand cubic feet) of gas injected. Another parameter considered is the gas–oil ratio (GOR), and this reflects the additional capacity needed at the wellsite to accommodate the increased gas production rate during the injection phase, thus preventing flaring. GOR is calculated at the conclusion of the injection phase. Based on the simulation results, it was observed that gas injection at a rate of 1 MMcfd (million cubic feet per day) through normal completion yielded the highest EOR efficiency, with a value of 0.18 bbl/Mcfd (barrel per thousand cubic feet). Additionally, up to 60% of the injected gas was successfully recovered during the process.

• Continuous injection of 2.8 MMcfd for three months:

Second case, a continuous injection of 2.8 MMcfd for three months. Two schemes were considered (Figure 6), injection through normal completion and injection through smart completion in which the lateral was segregated into three sections. As can be seen, both cases gave approximately the same incremental production. Relatively higher produced gas during the injection phase was observed in the case of normal completion. Higher oil and gas rates were obtained as soon as the injector was switched back to production.

The simulation results are shown in

Table 3. Simulation results of 2.8 MMcfd injection (3-month injection period).

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)
2.8 MMcfd Nor_Complet	72%	50.3	0.20	1.5
2.8 MMcfd Smart_Complet_ 3 Sections	71%	50.6	0.20	1.4

. Both approaches yielded the same gas injection efficiency and the same incremental oil and gas recovered. There was slightly higher GOR in the case of injection through normal completion. Up to 72% of the injected gas was recovered. Overall, there was a better injection efficiency and gas recovery compared to the previous EOR schemes.

• Continuous injection of 13 MMcfd for three months:

In this case, a dedicated injection of 13 MMcfd for three months was conducted. Again, injection was through normal completion and injection through smart completion in which the lateral was segregated into three sections. As shown in Figure 7, injection through normal completion yielded better incremental oil and higher gas production rates during the injection phase. Higher oil and gas rates were obtained as soon as the injector was switched back to production in both cases.

The simulation results are provided in

Table 4. Simulation results of 13 MMcfd injection (3-month injection period).

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)
13 MMcfd Nor_Complet	72%	162	0.18	3
13 MMcfd Smart_Complet 3 Sections	68%	180	0.21	1.9

. Gas injection through intelligent completion demonstrated the highest injection efficiency, resulting in greater incremental oil recovery. On the other hand, the injection through normal completion gave better gas recovery, up to 72% of the injected gas was recovered and higher GOR.

To summarize, using smart completion to inject the gas for a short period of time will not generate a decent increase in production. Based on the simulation results, 3 MMcfd or more is the optimum gas injection rate.

Figure 8 illustrates the bottom hole pressure simulated for 2.8 MMcfd and 13 MMcfd cases, respectively. As can be seen, BHP reached the maximum 7100 psi after 10 days of gas injection for 13 MMcfd injection case, while the maximum BHP obtained during gas injection of 2.8 MMcfd was 4500 psi.

• Continuous injection of 1 MMcfd for one year:

In this set of EOR schemes, gas injection of 1 MMcfd for one year was carried out (Figure 9). The injection through smart completion in which the lateral was segregated into three sections yielded a better incremental oil and clearly produced less gas. The oil rate reached the peak (700 bbl/d) after 10 months, which suggests the possibility of extending the injection phase to increase oil recovery.

Higher oil and gas rates were obtained as soon as the injector was switched back to production in both cases.

The simulation results are depicted in

Table 5. Simulation results for 1 MMcfd (one-year injection period).

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)
1 MMcfd Nor_Complet	70%	85	0.23	2
1 MMcfd Smart_Complet 3 Sections	60%	92.3	0.25	1.3

. Injection through intelligent completion yielded the highest gas injection efficiency (0.25 bbl/Mcf) and higher incremental oil with comparatively lower GOR. On the other hand, injection through normal completion gave better gas recovery, up to 70% of the injected gas was recovered.

• Continuous injection of 2.8 MMcfd for one year:

Gas injection of 2.8 MMcfd for one year was simulated. Three cases were considered (Figure 10)—the black line represents the base case (no gas injection), injection through normal completion, injection through smart completion in which the lateral was segregated into two sections, and injection through smart completion in which the lateral was segregated into three sections, starting the injection in the toe section, then middle section and finally the heel section. The results demonstrate that the injection through smart completion in which the lateral was segregated into three sections yielded the highest incremental oil with clearly produced less gas followed by injection through smart completion where the lateral was segregated into two sections. The oil rate reached the peak (1400 bbl/d) after 10 months, which suggests the possibility of extending the injection phase to increase oil recovery. A higher gas rate was obtained as soon as the injector was switched back to production in all three cases.

The simulation results are illustrated in

Table 6. Simulation results for the 2.8 MMcfd (one-year injection period).

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)
2.8 MMcfd Nor_Complet	74%	187	0.19	3.4
2.8 MMcfd Smart_Complet 2 Sections	65%	240	0.24	2.5
2.8 MMcfd Smart_Complet 3 Sections	60%	248	0.25	2.3

. Injection through intelligent completion in which the lateral was segregated into three sections yielded the highest gas injection efficiency (0.25 bbl/Mcf), higher incremental oil with comparatively lower GOR. On the other hand, injection through normal completion gave better gas recovery, up to 74% of the injected gas was recovered.

• Continuous injection of 13 MMcfd for one year:

Gas injection of 13 MMcfd for one year was carried out. Also three cases were considered as well (Figure 11). The injection through smart completion in which the lateral was segregated into three sections yielded the highest incremental oil with clearly produced less gas followed by injection through smart completion where the lateral was segregated into two sections. The oil rate reached the peak (2800 bbl/d) after 7 months. A higher gas rate was obtained as soon as the injector was switched back to production in all three cases.

The simulation results are presented in

Table 7. Simulation results for the 13 MMcfd one-year injection period.

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)
13 MMcfd Nor_Complet	82%	395	0.12	9.3
13 MMcfd Smart_Complet 2 Sections	80%	428	0.13	8.6
13 MMcfd Smart_Complet 3 Sections	80%	527	0.16	8.4

. Overall, lower gas injection efficiency and higher GOR were obtained. Injection through intelligent completion in which the lateral was segregated into three sections yielded the highest gas injection efficiency (0.16 bbl/Mcf), higher incremental oil with comparatively lower GOR.

To summarize, an injection rate of 2.8 MMcfd for one year yielded the highest gas injection efficiency with relatively manageable produced gas rate. However, the injection period can be optimized since the peak was reached after 10 months.

• Continuous injection of 2.8 MMcfd for three-year period:

In this case, gas injection of 2.8 MMcfd for three-year period was conducted, and three cases were considered (Figure 12). Injection through normal completion, injection through smart completion in which the lateral was segregated into three sections, and injection through smart completion in which the gas was injected through the middle section only. As can be seen, the injection through smart completion in which the lateral was segregated into three sections yielded the highest incremental oil. Relatively stable oil and gas rates were obtained in the case of injection through the middle section.

The simulation results are shown in

Table 8. Simulation results for the 2.8 MMcfd three-year injection period.

Case	Gas Recovered	Incremental Oil (Mbbl)	Injection Efficiency (bbl/Mcfd)	GOR (Mcf/bbl)	Incremental Oil Recovery
2.8 MMcfd Nor_Complet	72%	430	0.14	3.6	15.7%
2.8 MMcfd Smart_Complet 2 Sections	50%	488	0.16	2.3	17.9%
2.8 MMcfd Smart_Complet 3 Sections	67%	549	0.18	3	20.1%

. Injecting the gas for a long period of time will reduce the overall gas injection efficiency. Nevertheless, injecting the gas using smart completion and dividing the lateral into three sections yielded the highest incremental oil and highest injection efficiency. Forcing injection through the middle section has its benefits when injecting incremental oil and for moderate gas production rates, which makes the approach ideal if priority is given to less gas production.

Figure 13 Illustrates the streamline results of gas injection through the middle section of the lateral. As shown, forcing the injection through the middle section of the lateral will increase the swiping efficiency. Consequently, more oil is to be produced and potentially slows down the gas breakthrough.

Implementing gas injection in the Bakken will require obtaining a Class II permit. Operating companies must demonstrate that their wells comply with both federal and state regulations, including implementing effective measures to protect underground sources of drinking water. According to our simulations, lease-to-lease communication is highly likely. As a result, unless the adjacent leases are owned by the same company, the state will not approve gas injection due to mineral ownership rights.

5. Economics

Examining reservoir performance and assessing the related infrastructure costs is crucial to determining the economic feasibility of a project. In addition to these factors, there are several other costs that must be considered in the overall cost calculation. These include expenses related to compression units, lease fees, and miscellaneous costs associated with contracts, insurance, and monitoring.

Table 9. Smart completion and compression unit cost.

Item	Cost
Equipment	USD 600,000
Installation	USD 200,000
Engineering and Contingencies	USD 15,000 per month
Operation	USD 15,000 per month
Compressor Rental	USD 20,000 per month
Early Production Facility Rental	USD 35,000 per month
Intelligent Completion	USD 3,000,000

provides a depiction of the costs associated with purchasing compression units and the surface network. In an effort to minimize capital expenditure, it has been decided to rent the compressor skid instead of purchasing it, considering the relatively short injection period of the project. This decision is aimed at optimizing cost management and ensuring efficiency in the project’s financial planning. An early production facility is required to assist the expected increase in oil and gas rates during the injection phase.

This study utilized Internal Rate of Return (IRR), Levelized Cost of Energy (LCOE), and net present value (NPV) as key economic indicators. Two injection schemes were considered, injection of 2.8 MMcfd through smart completion for one year and 3-year periods. Sensitivity analyses on oil and gas prices were performed. In the first scenario, oil and gas prices were set at USD 80 per barrel and USD 8 per Mcf, respectively. In the second scenario, lower prices were evaluated, with oil at USD 50 per barrel and gas at USD 4 per Mcf.

Table 10. Compressor rental–revenue case of USD 80 per bbl and USD 8 per Mcf.

Case 1	Capital	OPEX	Incremental Oil	Revenue Oil	IRR	LCOE	NPV
2.8 MMcfd for one year	USD (3,800,000)	USD (4,200,000)	275,000	USD 22,000,000	129%	USD (32)	USD 9,400,000
2.8 MMcfd for three year	USD (3,800,000)	USD (11,000,000)	550,000	USD 44,000,000	285%	USD (27)	23,500,000

and

Table 11. Compressor rental–revenue case of USD 50 per bbl and USD 4 per Mcf.

Case 1	Capital	OPEX	Incremental Oil	Revenue Oil	IRR	LCOE	NPV
2.8 MMcfd for one year	USD (3,800,000)	USD (3,800,000)	275,000	USD 12,400,000	74%	USD (26)	USD 4,600,000
2.8 MMcfd for three year	USD (3,800,000)	USD (7,000,000)	550,000	USD 27,500,000	171%	USD (20)	13,000,000

illustrate, respectively, the cost and revenue for each case. OPEX calculated includes the cost of purchasing the gas.

An NPV of up to USD 23.5 million can be achieved with an IRR of 285% under a scenario involving 2.8 MMcfd gas injection over a three-year period. Even in a lower oil price environment of USD 50 per barrel, the project is projected to yield a profit with an NPV of USD 13 million.

6. Conclusions

Segregating laterals for gas re-injection to improve efficiency in shale plays using smart completion for enhanced oil recovery and flare reduction was investigated via numerical simulation. The developed reservoir model was history matched and was used to study and optimize the proposed methodology. The main conclusions of this study are as follows:

• Based on the streamline results, injecting the gas through normal completion has low sweeping efficiency. Approximately the first 20% of the perforated lateral is responsible for 80% of the injected gas, hence the lower oil volumes produced for each Mcf of gas injected.
• At least one-third of the lateral must be kept open in order to inject 3 MMcfd or more. In other words, segregating the lateral into four sections or more will not allow for a higher amount of gas injection.
• Injecting the gas in the last 30% of the lateral section (near the toe) has low incremental production and a significant surface compression requirement. The BHP may exceed 8000 psi for a gas injection rate of 2 MMcfd.
• The implementation of segregating laterals for gas re-injection, in conjunction with smart completion techniques, resulted in improved gas injection efficiency within shale plays, ranging from 0.18 to 0.25 bbl for each Mcf injected. This approach allowed for better management of gas injection and production, leading to enhanced oil recovery from the reservoir, so up to 20% incremental oil recovery can be achieved by injecting 2.8 MMcfd for three years.
• Incremental oil production from injection was greater in the case where the lateral was divided into three sections versus two sections. It is worth noting that segregating the lateral for injection has no benefit if the injection period is short (three months). Based on the simulation results, gas injection rates of 1MMcfd up to 3 MMcfd for one year have the highest gas injection efficiency.
• The results of this effort suggest that gas re-injection in the Bakken can be economically achievable. A broad estimation of the CAPEX and OPEX for a site designed to re-inject 2.8 MMcfd of gas through intelligent completion for three years is approximately USD 3.8 million and USD 11 million, respectively. The economic result obtained in terms of NPV at a discounted rate of 7% is a positive value, indicating the potential profitability of the project. Up to USD 23.5 million profit can be generated.