Selection of Enhanced Oil Recovery Method on the Basis of Clustering Wells

Abstract

The relevance of the technical and economic evaluation of the application of enhanced oil recovery methods at oil fields at the final stage of development is related to the need to recover the remaining reserves, including hard-to-recover (HTR) reserves, the share of which is growing annually. Currently, there are many effective enhanced oil recovery (EOR) methods for different process conditions, but their application has different effects based on the combination of methods, techniques and production conditions. The aim of this study was to approach the scaling of the effect of the application of modern EOR using the methodology of the clustering of wells with similar technological characteristics. This paper proposes a methodology for the selection of candidate wells to form clusters based on a set of indicators that determine the choice of enhanced oil recovery technology in oil fields at the final stage. The technological efficiency of sidetracking and multistage hydraulic fracturing application was evaluated based on the analytical method of well flow rate estimation. By applying cluster analysis to selected wells, three clusters were formed, each including three wells, united by the geological properties of their reservoir rocks and the filtration–capacitive properties of the oil. After this, the optimal technologies were selected for two clusters—hydraulic fracturing and sidetracking. The accumulated oil production, recovered due to the application of the technologies, from six wells for the first 7 years after the operation was estimated at 306.92 thousand tons of oil. Due to the achieved technological effect, the economic efficiency of the development of the studied oil field will increase due to the proceeds from the sales of the extracted additional oil. The results of this study can be used in the calculation of technical and economic efficiency at oil fields with similar conditions.

1. Introduction

According to worldwide and Russian energy development forecasts [1], both today and in 2050, the share of oil occupies a significant position in the structure of the worldwide energy consumption, so it is necessary to maintain a sufficient level of oil production [2,3,4]. However, a number of large oil fields are at the final stage of development, which implies the use of enhanced oil recovery (EOR) technologies to extract the remaining reserves, including hard-to-recover (HTR) reserves. According to Rosnedra (The Federal Agency for Subsoil Use, Russian Federation), the share of HTR reserves in 2023 was 58%, and it is expected to grow further, which also increases the demand for the development of geological and engineering operations (well interventions) [5].

The application of such types of well interventions as physical–chemical, hydrodynamic, bottomhole treatment such as acid treatment, the optimization of downhole pumping equipment, and perforation methods, as well as some types of hydraulic fracturing, is conditioned by the presence of unrecovered oil reserves in the drainage zone of the candidate well [6] in which these geological and engineering operations are planned to be carried out [7,8].

If there is a zone of unrecovered oil reserves in the interwell space, which belong to local intervals of the formation of weak depletion with the initial effective oil-saturated thickness, the most effective geological and technical measure will be sidetracking, which will have a point-like nature, allowing the prompt involvement of these remaining reserves in the development immediately after the sidetracks are put into operation. However, in addition to sidetracking, some form of hydraulic fracturing may also be successful.

The use of hydraulic fracturing as a method of increasing the oil recovery of certain low-permeability areas is one of the most effective technologies in the world [9], along with sidetracking. However, the process of creating artificial fractures, through which the flow of fluid to the wellbore will be carried out, requires a careful approach and accurate calculations, as well as the modeling of the process. Only with the combination of these conditions can the achievement of the greatest effect from the application of this measure be guaranteed [10]. In addition, after production from the stimulated zone of reserves, it is necessary to carry out repeated hydraulic fracturing [11,12]. Thus, it is necessary to carefully select a candidate well and compare the boundary criteria of the applicability of the technology with the geological conditions of the possible stimulated zones [13].

Therefore, it is necessary to focus on the development, modernization, and improvement of enhanced oil recovery methods, particularly hydraulic fracturing and sidetracking, in order to achieve more successful, cost-effective, and highly profitable results from the application of these technologies.

Taking into account the efficiency of these two methods of oil recovery, it is necessary to have a tool for their selection. In this paper, we propose the method of grouping wells by six attributes in order to cluster them and select the technology of enhanced oil recovery based on the parameters of these clusters. In the course of this study, the technological effect of sidetracking and multistage hydraulic fracturing was calculated.

The purpose of this study is to evaluate the application of enhanced oil recovery technologies in oil fields at the final stage of development for clusters of wells with specified characteristics. The objectives of the study were as follows: a review of the domestic and international experience in the use of technological solutions for enhanced oil recovery; the development of a methodology for the selection of candidate wells and the formation of clusters for the application of enhanced oil recovery technologies; the evaluation of the technological efficiency of sidetracking and multistage hydraulic fracturing on the basis of the cluster methodology.

The main idea of this study is that a certain method (e.g., cyclic flooding, chemical methods, sidetracking, hydraulic fracturing, etc.) is necessary to improve the efficiency of the development of hard-to-recover oil reserves at the final stage of production. However, the choice of method is conditioned by many technological factors related to the production and geological conditions of the field. To optimize the process of selecting an enhanced oil recovery method, an approach is proposed that consists of selecting wells using cluster analysis according to the given characteristics. In this study, based on the tree clustering method, three pools of wells with similar characteristics according to six attributes were formed, focusing on the well depth, density, oil viscosity, reservoir permeability, and reservoir pressure. For each cluster with a set of similar characteristics, it is possible to apply the most technologically effective method of oil recovery enhancement, which will provide the highest profitability in oil production.

2. Literature Review

2.1. Sidetracking Technology

Sidetracking is the process of creating a new wellbore by drilling a secondary one from the main string. Thanks to modern directional drilling technologies, the optimal utilization of existing wells, including those that are idle for various reasons, can be achieved. One of the most effective technologies today is the return to operation of idle wells, as well as increasing the productivity of wells with low flow rates through sidetracking [14,15]. This method is characterized by the fact that it is able to reconstruct the well exclusively along the necessary direction and is not limited to any depth or inclination angles. The greatest benefit of the sidetracking method is related to those producing fields that are at the final stage of development, where drilling new wells may not be profitable, as it will certainly not be possible to obtain a high flow rate.

Particular cases of sidetracking involve the creation of multilateral and other horizontally branched wells, which can bring the maximum economic effect, as they can simultaneously develop several disconnected sublayers and return to development the idle wells that have large remaining oil reserves in the drainage zone and are unable to return to operation when using other methods of EOR [16,17].

The widespread use of sidetracking demonstrates that, although sidetracking is complex, it is a very important technological process that allows one to significantly reduce the capital investments compared to drilling new wells, with the aim to infill the development system and to bring back into operation emergency wells with complications, low flow rates, and high water cuts. In addition, it allows one to involve so-called hard-to-recover (HTR) reserves [18] in development, which either have high viscosity or occur in low-permeability reservoirs. Thus, sidetracking is one of the most effective geological and engineering operations that influences the recovery of hydrocarbons with various geological properties in the reservoir at fields at the final stage of development [19].

The experience in the application of sidetracking technology will be considered through the example of the Yugra fields, due to the fact that this information is publicly available and that they belong to one of the main oil-and-gas bearing regions of the Russian Federation. As of the beginning of 2019, the number of producing wells with low flow rates is about 18,000, which is about 20% of all producing wells. In addition, about 7000 of the low-yielding wells are idle, i.e., non-operational, which amounts to approximately 9% of the total number of producing wells [14].

If we analyze the oil flow rates obtained from re-entered wells, we observe a decrease of 13 tons per day over ten years, namely from 42 tons per day in 2008 to 29 tons per day in 2018, i.e., by about 30% [14].

The dynamics of the number of sidetracking operations over a seven-year period, as presented in Figure 1, demonstrates that, during the period under consideration, the additional oil production increased by 1050, i.e., from 4450 to 5500 tons/year, which is 23%. Analyzing this graph, we also see that, for one sidetracking operation, there is a greater number of hydraulic fracturing operations at almost the same volume of additional oil production. Moreover, if we take into account only the beginning and the end of the studied interval, we observe that the volume of additional oil production obtained by hydraulic fracturing remains approximately at the same level, which proves the better efficiency of sidetracking in relation to hydraulic fracturing.

When analyzing the geological and engineering operations carried out to increase the oil recovery at the Solikamskoye field in Perm Krai, the highest increase in the oil well flow rate of 14 tons per day after sidetracking is observed, while the average increase among other methods is estimated at 10 t per day. Moreover, the application of sidetracking technology, in addition to covering the uninvolved oil reserves, eliminates the need to drill new wells and build other facilities and equipment for construction, which greatly reduces the capital investments in the project.

Figure 2 shows a visual graph of the EOR technologies at the Solikamsk fields, which reflects such operations as hydraulic fracturing, sidetracking, and bottomhole zone treatment. We see that sidetracking is the most effective method, which brings higher production growth and is actively used at these fields. During the ten-year period, about 170 well operations were carried out exclusively on wells that were in the well stock that was taken out of development [20].

If we analyze the cost of sidetracking, the capital investment required to construct a new well exceeds that of sidetracking by 30%, and, in some cases, even by 50%. In addition to the savings in carrying out the operation, the technology of sidetracking makes it possible to use the existing surface equipment and practically eliminates the cost of additional well pad construction.

From an environmental point of view, this technology also surpasses the construction of a new well.

At the Unvinskoye oil field, research on the parameters that potentially influence the initial well productivity index was carried out. The studied data were obtained from geophysical and hydrodynamic well studies. The filtration–capacitive properties of the rock and the physical and chemical properties of the fluid were also taken into account. As a result of the analysis, it was found that, at the initial well productivity index above 10, the flow rates of wells with sidetracks are most strongly influenced by such indicators as the length of the sidetrack and the size of the deviation angle when drilling it. Thus, by adjusting these parameters, it is possible to regulate the process of the extraction of the remaining reserves [20].

Along with the positive trend in carrying out such methods of EOR as sidetracking, the capital investments and complexity of their construction increase, as there is a need to involve the most remote and challenging remaining oil reserves, as well as the need for the long-term operation of the facility.

Thus, engineers face the challenges of constructing sidetracks under certain conditions, as well as the complications that arise in doing so and during production, which will be discussed in more detail in the next section.

Often, sidetracking is performed on wells that have been taken out of service due to water cuts or breakage. The distance from the wellhead to the proposed sidetracking location in such cases must be intact so that the working conditions are suitable. It is also necessary to thoroughly investigate the wells drilled nearby to avoid accidental wellbore intersection; for this reason, the following rules exist [21]:

• The direction of the sidetrack should eliminate the risk of intersection with neighboring wells;
• The sidetrack should have the curvature allowed by the technical drilling capabilities;
• The location at which sidetracking begins should be selected so that the horizontal end of the wellbore falls within the oil-saturated zone;
• The sidetrack should be drilled in the direction that the technical capabilities of the equipment allow, and its trajectory should coincide with the design trajectory as much as possible.

2.2. Hydraulic Fracturing Technology

Hydraulic fracturing is one of the most common methods of EOR used in hydrocarbon fields. The essence of the method is to create artificial fractures by injecting a fracturing fluid into the formation under high pressure; as a result of the formation of new fractures, it is possible to increase the drainage coverage, which will allow us to more fully develop the deposit. To prevent fracture closure under the influence of rock pressure, a specific granulometric material called a proppant is used [22,23].

After the primary hydraulic fracturing, the zones covered by the fracturing operation are exhausted over time, and, consequently, the production rate of some fractured wells begins to decline. It becomes relevant to carry out the operation with repeated hydraulic fracturing (refracturing), which allows us to involve those low-permeability zones that were not covered earlier [11,24]. Undoubtedly, this technology offers a cheaper and, at the same time, effective alternative to drilling new wells.

In order to consider the practical applicability of this technology, it is necessary to examine the global and domestic experience in the use of refracturing in fields with different characteristics.

For example, in the paper “Toward better hydraulic fracturing fluids and their application in energy production: A review of sustainable technologies and reduction of potential environmental impacts”, Lashun Thomas writes that, without the hydraulic fracturing of low-permeability formations, gas production from America’s fields would decline by as much as 60% over the next 10 years. An analysis conducted by the American Petroleum University showed that the predicted production decline is related to the lack of capabilities to drill and place wells into production in order to infill the well pattern and avoid a decline in production [25,26]. This means that, without impacting the existing oil clusters, which contain unrecovered reserves, further field development at the final stage will be characterized by strongly declining production [27,28], which will have a negative impact on oil and gas production from an economic point of view.

It is necessary to introduce the successful application of refracturing with fracturing fluid diverting technology by means of solid dispersed particles in the Haynesville field. The increase in oil production in some wells where refracturing was performed reached 200% compared to the average production level of the field [29]. For this technology to be successful, it is necessary to keep records of the technological parameters of hydraulic fracturing and forecast the production for the next few years, because the price of reagents is high and the technology is implemented under high pressure, so there is a probability of accidents and, therefore, of a negative effect in terms of the economy [10,30].

The experience of hydraulic fracturing in the Salym fields shows that refracturing with the use of a high-viscosity friction reducer has a positive correlation with increased oil production from the low-permeability Achimov reservoirs. However, studies have shown that not all reserves are economically profitable, and there is a rapid increase in water cuts in the initial years of exploitation [31,32]. Therefore, it is necessary to precisely adjust the boundary conditions regarding the applicability of refracturing; moreover, it is necessary to improve the operation itself—for example, to improve the particles of the diverting fluids [33,34] used in refracturing.

Gazprom’s success in applying the technology of refracturing with chemical diverters should not be overlooked. For example, such an operation was implemented in 2018 at several production wells in Slavneft-Megionneftegaz. This experience of refracturing was found to be the most successful among the other wells in the Gazprom fields [35,36]. Nevertheless, in addition to the positive effect, there were some problems—in particular, the probability of breakage initiation in the packer section, the high costs of the operation, and the need to use coiled tubing at such temperatures that can lead to accidents [12].

Thus, having analyzed the application experience, we can conclude that there are both positive and negative features that characterize the geological and engineering operation of refracturing. Successful refracturing will be achieved when the correct selection of a candidate well is carried out, the application parameters are determined in detail, and the modeling and forecasting of the expected fracture system is performed. Only in this case will oil recovery be increased due to the impact on the remaining oil reserves contained in the low-permeability sublayers.

When selecting candidate wells for geological and engineering operation application, certain indicators must be taken into account and the conditions described in

Table 1. Conditions for technology application (applicability criteria) (compiled by the authors on the basis of [21,37,38,39]).

Sidetracking Technology

Hydraulic Fracturing Technology

Remaining reserves—the sidetrack drainage zone should hold at least 10 thousand tons of oil. Oil production of the future sidetrack—not less than 5 tons per day. Net oil pay thickness—not less than 2.5 m; the smaller it is, the more difficult it is for the sidetrack to enter it. Formation pressure—injection wells ensure its support. Water cut—recommended for high-water-cut wells. Permeability—recommended for low-permeability reservoir rocks. Presence of an impermeable layer between oil- and water-saturated reservoirs. Sufficient distance from oil–water contact.

To utilize hydraulic fracturing, the following conditions must be satisfied [37]: Availability of recoverable reserves in the zone of the planned fracture—at least 6 thousand tons. Net oil pay thickness—recommended at least 5 m. Reservoirs consisting of terrigenous and carbonate rocks. Water cut—approximately 50–60%. Low permeability of reservoir rocks. High geological heterogeneity. Horizontal section of the well—for multistage hydraulic fracturing. Integrity of the production column and cement adhesion to it, especially in the area 50 m above and below the fracture location. Sufficient distance from oil–water contact.

must be met.

3. Materials and Methods

In the course of this study, we used open-access scientific publications, consisting of articles by domestic and foreign scientists, which highlighted theoretical and practical issues in the field of enhanced oil recovery methods.

The literature review included the study of the peculiarities of the application of enhanced oil recovery technologies via the example of an oil field in Western Siberia, which is at the final stage of development. The analysis of the reviewed information provides an opportunity to systematize the knowledge in the studied area and the possibility of identifying the features of these technologies.

The object of study was an oil field located in the Khanty-Mansi Autonomous Okrug region. It has been developed since the 1970s, and, today, it is at the fourth stage of development, i.e., the final stage. The reserve development of this field exceeds 80%. From the very beginning of its development, geological and engineering operations have been undertaken to improve the productivity rate of the production wells and the injection rate of the injection wells.

The results of the application of methods of enhanced oil recovery on the studied object are presented for 350 wells with additional oil production of 1647 thousand tons as a result of EOR application.

The selection of candidate wells for the study was performed by analyzing the well stock and information on the properties of the reservoir rocks and the fluids, which included a map of the current state of development and the oil-saturated thicknesses, as well as by comparing these data with previously specified criteria for the applicability of hydraulic fracturing and sidetracking. Nine wells were selected as potentially suitable for the implementation of the proposed methods.

The tree clustering method was applied for the study. This method involves the construction of a dendrogram using the Euclidean distance, which is a geometric distance described in a multidimensional space. The purpose of this method is to combine subjects into very large clusters using some measure of the similarity or distance between them.

Using the Statistica 10 software package, a distance matrix was obtained and then a dendrogram was modeled.

Based on the tree clustering method, 3 pools of wells with similar characteristics were formed according to 6 attributes: the net oil pay, well depth, oil density, dynamic viscosity of oil, reservoir permeability, and formation pressure.

The initial data were divided into 3 clusters by the K-means method, and the significance of the differences between the obtained groups was assessed. In the K-means method, the calculations start with K randomly selected observations (in this paper, K = 3), which become the centers of the groups, after which the object composition of the clusters is changed in order to minimize the variability within the clusters and maximize the variability between the clusters. After the cluster composition is changed, a new cluster center is computed, most often as a vector of averages for each parameter. The algorithm continues until the composition of the clusters no longer changes.

The calculation of the technological effect of sidetracking was carried out by the Joshi method [40] (

Table A1. Calculation of the technological effect of sidetracking operations [40].

Indicator	Formula
Calculation of flow rate without friction pressure losses (Joshi method) for a horizontal sidetrack	$Q_{HW}={\displaystyle \frac{2\pi K_h{h}_e\rho (P_f-P_b)}{\mu (ln\left(\frac{a+\sqrt{a^2-{\left(\frac{L}{2}\right)}^2}}{\frac{L}{2}}\right)+\frac{I_{ani}{h}_e}{L}·\mathit{ln}\left(I_{ani}\frac{h_e}{2r_w}\right)+S)}}$,	(A1)
	where Iani—anisotropy of horizontal and vertical permeability; a—major semi-axis of the drainage ellipsoid formed by the horizontal well, m; L—length of the horizontal section of the wellbore, m; Kv—vertical permeability of the target object in the sidetracking zone, m2; Kh—horizontal permeability of the target object in the sidetracking zone, m2; he—net oil pay thickness, m; ρ—oil density, kg/m3; μ—fluid viscosity, Pa·s; Pf—formation pressure, Pa; Pb—bottomhole pressure, Pa; rw—borehole radius, m; S—skin factor.
Anisotropy of horizontal and vertical permeability	$I_{ani}=\sqrt{{\displaystyle \frac{{\mathrm{К}}_h}{{\mathrm{К}}_v}}}$	(A2)
Major semi-axis of the drainage ellipsoid	$a={\displaystyle \frac{L}{2}}(\sqrt{0.5+{(0.25+{\left({\displaystyle \frac{2·{\mathrm{r}}_e}{L}}\right)}^4)}^{0.5}})$	(A3)
	where re—supply contour radius, m.
Friction pressure losses in the horizontal section of the well (Darcy–Weisbach formula)	$P_{friction}=\lambda \cdot {\displaystyle \frac{L}{r_w^5}}\cdot {\displaystyle \frac{Q^2}{64\cdot {\pi }^2}}\cdot 10^6\cdot {\rho }_{fluid}$	(A4)
	where λ—hydraulic resistance factor; rw—borehole radius, m.
Reynolds number (Re)	$\mathit{Re}={\displaystyle \frac{2\cdot r_w\cdot Q\cdot {\rho }_{fluid}}{\mu \cdot \pi \cdot {r_w}^2}}$	(A5)
Hydraulic resistance factor	$\lambda ={\displaystyle \frac{64}{Re}}$	(A6)
Determination of oil flow rate	$Q_{oil}=Q_{HW}\ast (1-{\mathrm{B}}_i)$	(A7)
	where Bi—water cut, %.
Calculation of indicators for Well 1	Major semi-axis of the drainage ellipsoid: $a={\displaystyle \frac{50}{2}}(\sqrt{0.5+{\left(0.25+{\left({\displaystyle \frac{2·500}{\frac{50}{2}}}\right)}^4\right)}^{0.5}})=500.31\mathrm{m}$ Anisotropy of permeability: $I_{ani}=\sqrt{{\displaystyle \frac{40.08}{1.84}}}=4.67$ Expected flow rate of a horizontal wellbore at its length of 50 m: $Q_{HW}={\displaystyle \frac{2\mathsf{\pi }·40.08·{10}^{-15}·3.1·876·(24.5-17.2){·10}^6}{1.42·{10}^{-3}·(\mathrm{l}\mathrm{n}\left(\frac{500.31+\sqrt{{500.31}^2-{\left(\frac{50}{2}\right)}^2}}{\frac{50}{2}}\right)+\frac{4.67·3.1}{50}·\ln \left(\frac{4.67·3.1}{2·0.1}\right)+1.1)}}=0.000665{\displaystyle \frac{{\mathrm{m}}^3}{\mathrm{s}}}$ Reynolds number: ${Re}_1={\displaystyle \frac{2·0.1·0.876·0.665·{10}^{-3}}{1.42·{10}^{-3}·\pi ·0.1}}=\mathrm{20,615}$ Hydraulic resistance factor: ${\lambda }_1={\displaystyle \frac{64}{2.615}}=24.48$ Friction pressure losses: ${∆\mathrm{P}}_{friction}=24.48·{\displaystyle \frac{50·{\left(0.665·{10}^{-3}\right)}^2}{64·{0.1}^5·{3.14}^2}}·0.867=\mathrm{75,236}\mathrm{Pa}$ Liquid flow rate taking hydraulic losses into account: $Q_{HW}={\displaystyle \frac{2\mathsf{\pi }·40.08·{10}^{-15}·3.1·876·(24.5-17.2-0.075){·10}^6}{1.42·{10}^{-3}·(\mathrm{l}\mathrm{n}\left(\frac{500.31+\sqrt{{500.31}^2-{\left(\frac{50}{2}\right)}^2}}{\frac{50}{2}}\right)+\frac{4.67·3.1}{50}·\ln \left(\frac{4.67·3.1}{2·0.1}\right)+1.1)}}=0.000658{\displaystyle \frac{{\mathrm{m}}^3}{\mathrm{s}}}$ Oil flow rate of Well 1 at the sidetrack length of 400 m: $Q_{oil1}=87.58-\left(1-0.738\right)=22.92\mathrm{ton}/\mathrm{day}$

and

Table A2. Well flow rates depending on the length of the sidetracks.

L, m	Well 1		Well 6		Well 8
	Qf, m3/Day	Qf Taking into Account the Losses, m3/day	Qf, m3/day	Qf Taking into Account the Losses, m3/day	Qf, m3/day	Qf Taking into Account the Losses, m3/day
50	57.50	56.90	52.18	51.72	41.55	41.25
100	73.50	71.57	66.18	64.68	56.47	55.34
150	84.46	80.63	76.03	73.05	67.25	64.86
200	93.36	87.11	84.17	79.31	76.34	72.23
250	101.12	91.96	91.39	84.22	84.53	78.22
300	108.17	95.59	98.02	88.13	92.16	83.16
350	114.74	98.22	104.26	91.21	99.41	87.19
400	120.96	99.98	110.20	93.54	106.35	90.37
450	126.92	100.93	115.92	95.18	113.04	92.72
500	132.68	101.12	121.46	96.16	119.45	94.24
550	138.27	100.57	126.82	96.48	125.60	94.94

, Appendix A) and in case of multistage hydraulic fracturing using the Mishchenko method [41] (

Table A3. Calculation of the technological effect of hydraulic fracturing according to Mishchenko’s methodology [41,71].

Indicator	Formula
Vertical component of rock pressure	$P_{v.r.}={\rho }_{rock}·g·L$	(A8)
	where L—well depth, m; ρrock—rock density, kg/m3; g—gravity acceleration, m/s2.
Horizontal component of rock pressure	$P_{h.r.}={\displaystyle \frac{v}{1-v}}·\left(P_{v.r.}-P_f\right)+P_f$	(A9)
	where v—Poison’s ratio, unit fractions; Pf—formation pressure, Pa.
Hydraulic fracturing pressure at the well bottomhole	$P_{hf.bth.1}=P_{v.r.}-P_f+G_t$	(A10)
	where Gt—resistance of rock to tearing, assumed to be equal to 2.24 MPa.
Required bottomhole pressure	$P_b=P_{hf.bth.}·a$	(A11)
	where a—is the excess of the bottomhole pressure over the hydraulic fracturing pressure, equal to 1.2.
Volume concentration of sand in the mixture	${\beta }_s={\displaystyle \frac{\raisebox{1ex}{$C_{sand}$}\!\left/ \!\raisebox{-1ex}{${\rho }_{sand}$}\right.}{\left(\raisebox{1ex}{$C_{sand}$}\!\left/ \!\raisebox{-1ex}{${\rho }_{sand}$}\right.\right)+1}}$	(A12)
	where Csand—sand concentration per m3 of liquid, kg/m3; ρsand—sand density, kg/m3.
Density of the mixture injected into the reservoir	${\rho }_m=\left({\rho }_{sand}-{\rho }_{oil}\right)·{\beta }_s+{\rho }_{oil}$	(A13)
	where ρoil—oil density, kg/m3.
Static pressure in the well	$P_{st}={\rho }_m·g·L$	(A14)
Mixture viscosity	${\mu }_m={\mu }_f·e^{3.18·{\beta }_s}$	(A15)
	where μf—fluid viscosity, Pa·s.
Reynolds number	$Re={\displaystyle \frac{4·Q·{\rho }_m}{\pi ·d·{\mu }_m}}$	(A16)
	where Q—injection rate, m3/s; d—wellbore diameter, m.
Hydraulic resistance factor	$\lambda ={\displaystyle \frac{0.3164}{{Re}^{0.25}}}$	(A17)
Estimation of friction pressure losses	$P_{friction}=1.52·\lambda ·{\displaystyle \frac{16·Q^2·L·h}{6.28·d^5}}$	(A18)
	where h—net oil pay thickness, m.
Wellhead pressure at hydraulic fracturing	$P_{wh}=P_b-P_{st}+P_{friction}$	(A19)
Determination of optimal fracture half-length	$x_{frac}={\displaystyle \frac{143}{{\left(1000·k_f\right)}^{0.27}}}$	(A20)
	where kf—formation permeability, D.
Fracture wing area	$S_{frac}=2·h·x_{frac}$	(A21)
Proppant weight	$M_{sand}={\displaystyle \frac{i_{sand}·S_{frac}}{1000}}$	(A22)
	where isand—proppant distribution coefficient in the formed fracture, equal to 4.5 kg/m2.
Sand carrying fluid column	$V_{sandf.}={\displaystyle \frac{M_{sand}}{{\mathrm{C}}_{sand}}}$	(A23)
Displacement fluid volume	$V_{dispf.}={\displaystyle \frac{1.3·\pi ·d^2·L}{4}}$	(A24)
Calculation of the total volume of injected fluid	$V_{injf.}=V_{fracf.}+V_{sandf.}+V_{dispf.}$	(A25)
	where Vfrac f.—volume of the fracturing fluid, equal to 1/9 of $V_{sandf.}$
Duration of hydraulic fracturing operation	$t={\displaystyle \frac{V_{injf.}}{Q}}$	(A26)
Number of pumping units	$N={\displaystyle \frac{P_{wh}·0.011}{0.8·36·Q}}+1$	(A27)
Fracture length	$l=\sqrt{{\displaystyle \frac{V_{injf.}·E}{5.61·h·\left(1-{\lambda }^2\right)·\left(P_b-P_{h.r.}\right)}}}$	(A28)
	where E—modulus of elasticity of rocks, Pa.
Fracture width	$\omega ={\displaystyle \frac{4·l·\left(1-v^2\right)·\left(P_b-P_{h.r.}\right)}{E}}$	(A29)
Fracture volume	$V_{frac}=2·h·x_{frac}·\omega$	(A30)
Fracture width after healing	${\omega }^{\prime }={\displaystyle \frac{\omega ·{\beta }_s}{1-m}}$	(A31)
	where m—porosity, %.
Permeability of a fracture as a function of its width	$k_{frac}^{\prime }={\displaystyle \frac{{\omega \prime }^2}{12}}$	(A32)
Permeability (Carman–Kozeny)	$k_{frac}={\displaystyle \frac{m^3}{36·C·{\left(1-m\right)}^2}}·d^2$	(A33)
Average permeability value	$K_{frac}={\displaystyle \frac{K_{frac}^{\prime }+k_{frac}}{2}}$	(A34)
Pressure at the boundary of the interfracture space	$p_0={\displaystyle \frac{P_f-\left(\frac{1}{2}-{\left(N-1\right)}^{2}A\right)·P_b}{\frac{1}{2}+{\left(N-1\right)}^{2}A}}$	(A35)
	where N—number of stages of fracturing, units.
Interfracture space density	$A={\displaystyle \frac{2·x_{frac}·\left(r_e-x_{frac}\right)}{{L_{HW}}^2}}$	(A36)
Flow rate for 4 fractures	$Q_{IV}={\displaystyle \frac{2·k_f·h·L_{HW}}{{\mu }_{oil}·\left(r_e-x_{frac}\right)}}·\left(P_f-{\displaystyle \frac{p_0}{2}}-{\displaystyle \frac{P_b}{2}}\right)$	(A37)
Flow rate of the fifth fracture (Dupuis formula for radial inflow)	$Q_D={\displaystyle \frac{2·\pi {·k}_f·h}{{\mu }_{oil}·ln\left(\frac{r_e}{r_w}\right)}}·\left(P_f-P_b\right)$	(A38)
Determination of flow rate	$Q=Q_{IV}+Q_D$	(A39)

and

Table A4. Results of the technological effect of hydraulic fracturing.

Name of Indicator	Well 3	Well 4	Well 7
Flow rate of four fractures, ton/day	40.04	37.14	41.09
Flow rate of the fifth fracture, ton/day	17.91	19.07	16.77
Total fluid flow rate, ton/day	57.95	56.21	57.87
Water cut, %	74	69.7	72.4
Total oil flow rate, ton/day	16.32	17.59	17.09
Number of pump units, units	2	2	2
Sand weight, kg	8387	9984	8935
Operation time, min	72.5	80.9	75.5

, Appendix A). The overall study design is shown in Stages 1–4.

• Stage 1. Analysis of scientific literature

Method and Materials: Scientific articles and industry reports in the field of enhanced oil recovery methods and technical performance evaluation techniques.

• Stage 2. Analysis of efficiency of EOR technologies

Method and Materials: Experiments on EOR application in 350 wells in the study area were carried out. The following methods were applied:

-

Chemical (oil displacement using aqueous surfactant solutions, polymer and alkaline solutions, acid treatment);

-

Hydrodynamic (flow-diverting technologies and cyclic flooding);

-

Physical (hydraulic fracturing and its types, sidetracking).

Flow rate increases were determined (for results, see Stage 3 (Result)).

• Stage 3. Development of methodology for selection of candidate wells and formation of clusters for application of enhanced oil recovery technologies

Method and Materials: Identification of well characteristics (attributes) that correspond to the application of effective EOR (for results, see

Table 2. Characteristics of selected wells.

	Net Oil Pay, m	Depth, m	Oil Density, kg/m3	Dynamic Viscosity of Oil, mPa·s	Permeability, mD	Formation Pressure, mPa
Well 1	3.1	2358	876	1.42	1.84	24.5
Well 2	9.2	3680	919	19.3	10.9	28.3
Well 3	14.2	2800	884	10.3	17.9	26.5
Well 4	16.4	2852	894	10.1	16	26.5
Well 5	14.4	3560	929	20.1	11	28.3
Well 6	2.8	2262	873	1.34	2.03	24.5
Well 7	14.8	2826	889	10.7	16.5	26.5
Well 8	4.2	2366	870	1.39	1.04	24.5
Well 9	11.8	3400	924	19.7	9.5	28.3

);

Formation of clusters for application of EOR methods on basis of tree-shaped clustering technique with help of Statistica 10 software package (tree-shaped clustering method and dendrogram construction using Euclidean distance and K-means method were used).

Result: Three pools of wells with similar characteristics were formed according to six attributes: net oil pay, well depth, oil density, dynamic viscosity of oil, reservoir permeability, and formation pressure.

• Stage 4. Assessment of technological efficiency of sidetracking and multistage hydraulic fracturing application

Method and Materials: Calculation of technological effect of sidetracking was carried out by Joshi method [8] (Table A1 and Table A2, Appendix A) and, in case of multistage hydraulic fracturing, using Mishchenko method [41] (Table A3 and Table A4, Appendix A).

4. Results

The results for the studied object are presented for 350 wells with additional oil production of 1647 thousand tons as a result of the application of EOR; see Figure 3.

Having analyzed the above information, we can conclude that the most effective technologies for enhanced oil recovery at the studied field are hydraulic fracturing and sidetracking. The average oil flow rate of wells equipped with sidetracks reaches 26.5 t per day, and that at multistage hydraulic fractured wells reaches 33.4 t per day.

The selection of candidate wells for the studied object was performed by analyzing the well stock and information on the properties of the reservoir rocks and the fluids, which included a map of the current state of development and the oil-saturated thicknesses, as well as by comparing these data with previously specified criteria for the applicability of hydraulic fracturing and sidetracking. Nine wells were selected as potentially suitable for the implementation of the proposed methods, and their characteristics are listed in Table 2.

Analyzing the presented data, we can see the differences in each of the parameters. It is necessary to further group the selected wells; for this purpose, we conduct a cluster analysis using the six parameters listed in Table 2. For the wells of the formed clusters, we subsequently select the necessary enhanced oil recovery technologies.

For this study, it is appropriate to firstly apply the tree-shaped clustering method. Using the program “Statistica 10”, the distance matrix was obtained (

Table 3. Distance matrix.

	Well 1	Well 2	Well 3	Well 4	Well 5	Well 6	Well 7	Well 8	Well 9
Well 1	0	1323	443	495	1203	96	469	10	1043
Well 2	1323	0	881	828	121	1419	855	1315	280
Well 3	443	881	0	53	761	539	27	435	601
Well 4	495	828	53	0	709	591	27	487	549
Well 5	1203	121	761	709	0	1299	735	1196	160
Well 6	96	1419	539	591	1299	0	565	104	1139
Well 7	469	855	27	27	735	565	0	461	575
Well 8	10	1315	435	487	1196	104	461	0	1036
Well 9	1043	280	601	549	160	1139	575	1036	0

), and then the dendrogram was modeled (Figure 4) from the data, as presented in Table 2.

4.1. Formation of Clusters for Application of EOR Methods

Based on the tree clustering method, three pools of wells with similar characteristics were formed using six attributes: the net oil pay, well depth, oil density, dynamic viscosity of oil, reservoir permeability, and formation pressure.

The X-axis shows the wells and the Y-axis shows the relative distance between them, i.e., the Euclidean distance.

Based on the obtained dendrogram, we can assume that the objects form three natural clusters. Let us verify this assumption by dividing the initial data, using the K-means method, into three clusters and check the significance of the differences between the obtained groups.

Table 4. Variance analysis.

	Between	cc	Within	cc	F	Significance
Net oil pay, m (X1)	221	2	17.19	6	38.507	0.000378
Depth, m (X2)	2,250,225	2	47,517.47	6	142.067	0.000009
Oil density, kg/m3 (X3)	4082	2	118.00	6	103.780	0.000022
Dynamic viscosity of oil, mPa·s (X4)	503	2	0.51	6	2961.098	0.000000
Permeability, mD (X5)	348	2	3.90	6	267.784	0.000001
Formation pressure, mPa (X6)	22	2	0.00	6	155.3695	0.000007

shows the results of the variance analysis to determine the significance of the differences between the obtained clusters; the value of p < 0.05 indicates a significant difference. Based on the obtained data, we can conclude that all factors have sufficient significance. Figure 5 shows a graph of the K-means results.

Table 5. Clustering by K-means method with distances to the centers of the clusters.

Cluster 1			Cluster 2			Cluster 3
Well 3	Well 4	Well 7	Well 2	Well 5	Well 9	Well 1	Well 6	Well 8
10.825	10.826	0.228	54.482	5.916	59.878	12.039	27.218	15.296

shows the clustering by the K-means method, from which we can see an identical pattern regarding the results of the hierarchical method above, indicating that the formed grouping is valid.

After considering the characteristics of the wells belonging to the first cluster and comparing them with the map of oil-saturated thicknesses, it was decided that multistage hydraulic fracturing would be more effective as the applicability criteria mentioned above are met. Wells 3, 4, and 7 are equipped with horizontal completion and are low-production wells (up to 5 t per day) due to the low permeability and high water cut.

After studying the characteristics of the wells belonging to the third cluster and comparing them with the map of the current state of development and the oil-saturated thicknesses, it was decided to perform sidetracking, which will provide high efficiency in terms of increasing oil production. The above conditions of applicability for this case are also met. Wells 1, 6, and 8 are currently suspended due to the high water cut, but there are sufficient remaining reserves nearby that can be brought into development using sidetracking.

The wells belonging to the second cluster were also analyzed on the basis of their characteristics and compared with the maps of the current state of oil-saturated strata development, but it was deduced that the necessary technological effect from the application of hydraulic fracturing would not be achieved, which was confirmed by the calculations performed later (the average fluid flow rate of the three wells was 13.06 t per day, and the average oil flow rate was 3.89 t per day).

4.2. Estimation of Technological Effect

The calculation of the technological effect of sidetracking was carried out by the Joshi method [40] (Table A1 and Table A2, Appendix A) and in case of multistage hydraulic fracturing using the Mishchenko method [41] (Table A3 and Table A4, Appendix A).

The sidetrack lengths for Wells 1, 6, and 8 are 400 m, 500 m, and 450 m, respectively. Hydraulic fracturing was carried out in five stages. The results of the oil flow rate after the application of the studied technologies are shown in

Table 6. Oil flow rate results of the studied wells.

Indicator	Sidetracking			Hydraulic fracturing
	Well 1	Well 6	Well 8	Well 3	Well 4	Well 7
Oil flow rate, tons per day	22.92	25.44	22.22	16.33	17.59	17.10
Average oil flow rate, tons per day	23.53			17.01
Average oil flow rate for the oil field, tons per day	26.5			33.4

.

Since, at the oil field, the average flow rate of wells equipped with sidetracks is estimated at 26.5 t per day, while the average flow rate of the wells under study (Wells 1, 6, and 8), resulting from these calculations, is 23.53 t per day, we can consider this experience a success, as the results are close to the real statistics. Regarding the flow rates of the wells that were fractured, the outcomes are similar: according to the statistics for the oil field, the average oil flow rate with hydraulic fracturing is 33.4 t per day, while the studied wells showed 17.06 t per day, which, although two times lower, can be considered successful because it is within the range of the flow rates for most of the producing wells.

The cumulative oil production by year was calculated, and the results are presented in

Table 7. Cumulative oil production by year.

Year	Sidetracking, Oil Production, Tons			Hydraulic Fracturing, Oil Production, Tons
	Well 1	Well 6	Well 8	Well 3	Well 4	Well 7
1	8240.81	9145.57	7989.33	5869.35	6325.82	6147.40
2	7816.60	8674.80	7578.07	5615.02	6051.70	5881.01
3	7533.42	8360.52	7303.53	5429.52	5851.77	5686.73
4	7288.05	8088.21	7065.64	5275.93	5686.25	5525.87
5	7080.48	7857.86	6864.41	5154.27	5555.12	5398.44
6	6910.73	7669.47	6699.84	5064.53	5458.40	5304.44
7	6778.79	7523.04	6571.93	5006.70	5396.07	5243.88
8	-	-	-	4981.00	5368.37	5216.96
9	-	-	-	4923.01	5305.87	5156.22
Cumulative	51,648.89	57,319.47	50,072.74	47,319.33	50,999.37	49,560.95
Total	159,041.1			147,880

. The remaining reserves in the area of wells equipped with sidetracks are planned to be recovered in 7 years, and those in the area of wells with hydraulic fracturing in 9 years (Table A2 and Table A4, Appendix A).

5. Discussion

To accurately determine the reservoir areas that will be the most effective for the application of sidetracking technology, it is necessary to use actual geological and hydrodynamic models of the selected field. To assess the potential efficiency, it is necessary to act according to the following plan [5,20]:

• Identify emergency wells with a high water cut and low flow rates that can only be improved by sidetracking;
• Assess the degree of oil recovery in the zones belonging to the candidate wells;
• Determine the choice of sidetracking point and the direction of its horizontal part and justify the choice of intervals planned for penetration;
• Evaluate the future impact, i.e., the interference, of the sidetrack on the performance of other wells that have penetrated the same reservoir;
• Evaluate the sidetracking operation and following production of the sidetrack in terms of economics.

When choosing the direction of sidetracking, it is necessary to take into account the filtration–capacitive properties of the formation, the distance from the oil–water contact, and the existence of a layer between the oil- and water-saturated parts of the formation. A vertically inclined direction of the sidetrack will work more effectively in deposits with seamless oil sections, high permeability, and weak water flooding. A horizontal direction is preferable in the zone of remaining reserves if it is at least 3 m away from the water– and gas–oil contact, as well as in deposits with high water flooding [14,20].

In order to eliminate the divergence probability problem identified in the previous section, it is necessary to perform a thorough preparatory study, including the construction of a modified wellbore profile using a gyroscopic inclinometer and an assessment of the technical condition of the casing [42,43]. The negative effects of pressure pulsation can be eliminated by selecting a suitable drilling fluid and adjusting its density to isolate potentially risky areas. In the case of problems that arise at the first drilling attempt, corrective measures must be taken, which will affect the cost of the operation. In order to eliminate the dependence on pulse equipment, it is necessary to stimulate its development by domestic manufacturers.

Despite the demanding preliminary preparation, the high cost of the technology relative to other EOR methods, and the potential risks, sidetracking is a popular means to recover remaining oil reserves, so it is necessary to further study and modernize the technology to make it even more effective and reduce the risks of complications [44,45,46].

Promising solutions to hydraulic fracturing problems have been outlined. Both foreign and domestic specialists have studied the experience of refracturing and identified parameters that can improve the efficiency of this technology [47,48]. In the article “Study on the mechanisms of refracturing technology featuring temporary plug for fracturing fluid diversion in tight sandstone reservoirs”, Li Wei considers the rock stress parameter during hydraulic fracturing and concludes that the best effect in the formation of new fractures will be achieved when there is a difference in the horizontal stresses in the affected zone [49]. Moreover, during the study, it was found that, in the case of fracture fluid injection at a low speed, the best effect of temporary plug formation is achieved; this, in conjunction with natural fractures, will reorient the newly formed fractures, thus forming a complex network of fractures, which will effectively engage the remaining oil reserves [49].

In the article “Applications of self-degradable particulate diverters in wellbore stimulations…”, J. Huang discusses refracturing with the use of a temporary plug to divert the fracture fluid. The essence of this method is to block the already formed fractures that contain proppant, preventing them from closing under rock pressure. Particles of the isolating compound, with the help of the fracture fluid, enter the formed fractures and temporarily seal them, as a result of which the pressure of the injection fluid increases [50,51]. This ensures the planned trajectory of new cracks and blocks the expansion of the initial cracks [52,53].

The practical use of this method has shown that temporary plugging successfully forms a new fracture network and covers the remaining oil reserves in the development, i.e., it is a successful method for enhanced oil recovery [50]. For example, at the Daqing oil field, this technology was tested in a low-permeability reservoir characterized by high heterogeneity. As a result of primary hydraulic fracturing, the oil production per unit of productive strata increased by 0.09 t/(day·m), and the accumulated production for 300 days was about 1050 t. After the application of refracturing with the use of temporary plugging, a large number of new fractures were formed, due to which the flow rate increased by 0.16 t/(day·m), and the accumulated oil production amounted to about 3400 t. Thus, the method of refracturing proved to be more successful in relation to the primary hydraulic fracturing [49]. To eliminate the risks of complications during operation, it is necessary to take into account the geological characteristics of the reservoir, the parameters of fracturing fluid injection, the injection pressure, the wellbore section, and the distance from other wells, as well as to use modeling technologies [54,55].

A cross-linked gel is considered to be a promising material in the isolation solution used for the temporary blocking of primary cracks [9,56]. A cross-linked gel based on polylactic acid is characterized as the most successful when used as a chemical diverter. The advantage of this material is that it is biodegradable, which eliminates the need for a degradation agent to remove it after refracturing [57,58]. Moreover, the composition of the isolation solution can be varied depending on the thermobaric conditions [59]. Due to the wide range of solution variations, it is possible to reduce the cost of the operation [9,60].

In order to eliminate the problems associated with the destruction of the proppant integrity, as well as the reduction in pressure in the crack, leading to the impact of the injected fracture fluid only in existing cracks, it was proposed to use a low-molecular composition based on polyacrylamide gels as diverters [61]. Their application sets the required pressure, providing the redistribution of stresses, and cracks are created in the planned direction; in addition, polyacrylamide gels are of high strength and heat-resistant, and these properties can expand the application of this technology [9,61].

However, the disadvantage of polyacrylamide gels is the high price of some components included in their compositions, so the profitability of this technology remains in question [54,62]. Therefore, it is necessary to conduct research on components with similar properties for the cross-linking process, but with a lower price. The improvement of this process will improve hydraulic fracturing technology, making it more economically and technologically efficient.

If we refer to the experience with hydraulic fracturing, particularly refracturing in Russian fields, we can see that a large number of operations do not achieve the expected effect due to their complexity. The incompleteness of the effect is manifested due to complications arising both during the geological and engineering operation and during the production of the treated well. One of the most common problems encountered by specialists during the operation of refracturing is the premature growth of the water cut of the produced fluid due to water breakthrough from highly water-saturated formations, which are not taken into account during the planning and modeling of the operation, as well as the inappropriate choice of the potential stimulated area [57,63]. Another highly common problem is the enlargement of the fractures already created during primary hydraulic fracturing and hence the absence of new fractures due to the impact applied on the areas of least resistance, i.e., old fractures, resulting in increased water cut growth [64,65]. This clearly shows that when creating a plan for refracturing, it is necessary to take into account the exact location of the already created primary fractures. For example, the locations of fractures can be detected by microseismic and high-frequency monitoring [66,67].

Sidetracking technology is devoid of these problems. As was noted earlier, sidetracking allows us to affect individual oil-saturated clusters in a much smaller, more localized area of the development site, unlike, for example, such EOR methods as fluid-diverting technologies, which are used to improve the displacement process and to evenly develop the remaining oil reserves over the entire area of the site. Sidetracking allows us to reduce the number of idle and high-water-cut wells, which, in general, are often potential candidate wells for this technology, as other geological and engineering operations that focus on enhancing oil recovery are not always able to manage this situation effectively [68,69,70].

However, engineers also encounter complications during sidetracking. The most common complications include the sidetracking itself, as well as the sidetrack casing process, which can be related to poor casing drifting, contamination, a lack of cement behind the casing in the window cutout area, the cavernous nature of the borehole, and milling mode interruptions.

The main limitation of this study is that, in determining the technological effect of the proposed solution, analytical calculations should be supported by the results of hydrodynamic modeling in specialized software for a more reliable result, as, otherwise, the calculation error may be high.

The value of this study lies in the methodology of grouping wells with similar characteristics using cluster analysis to select suitable enhanced oil recovery technologies. Taking into account the results of the application of enhanced oil recovery methods based on the example of the studied 350 wells, it should be recognized that, without the use of enhanced oil recovery measures, many wells would be unprofitable. Wells 1, 6, and 8 would have been suspended and idle with zero production. It is also difficult to determine, due to insufficient data, the production of Wells 3, 4, and 7 without the application of enhanced oil recovery. In other words, the application of the proposed technologies (hydraulic fracturing and sidetracking) and the clustering of wells according to their technological characteristics allows us to increase the profitability of the field’s development as a whole. The theoretical significance of this study is its assessment of the technical and economic effects of the application of the considered methods, with certain well characteristics. The results of Section 2 provide the basis for the selection of enhanced oil recovery technologies. The practical significance of this study lies in the fact that the proposed methodology can be used for producing wells in oil fields with similar characteristics. The limitations of this study include the use of limited data to predict the well production rates under the Joshi methodology (for sidetracking) and the Mishchenko methodology (for hydraulic fracturing); the use of secondary data, without building a hydrodynamic reservoir model, to identify the well characteristics for further clustering; the fact that the production forecast was calculated using the Joshi method (for sidetracking) and the Mishchenko method (for hydraulic fracturing) and the uncertainty was not considered separately; and the fact that confidence intervals and error bars were not considered.

The proposed approach evaluates the production, sales revenue, and capital and operating costs and considers several taxation options and, accordingly, several values of income for the subsoil user and the state. The proposed solution has a number of advantages; for example, since the technologies are applied to existing wells, the capital costs are lower. This study confirms that the application of enhanced oil recovery methods for oil fields at the final stage of development is feasible from a technical and economic point of view.

6. Conclusions

This paper analyzes the development of one of the oil fields in Western Siberia and considers the possibility of introducing technologies that enhance oil recovery, thus making the development of the field more cost-effective. The cluster analysis of the well characteristics was carried out to select certain technologies, the technological effect was calculated using analytical methods, and the economic efficiency of the selected solutions was estimated. The results of this study can be used in the calculation of technical and economic efficiency at oil fields with similar conditions.

For this study, we selected an oil field in Western Siberia, which is at the final stage and has hard-to-recover reserves, and analyzed the results of the application of enhanced oil recovery technologies. After this, we selected the most effective ones for these conditions, namely hydraulic fracturing and sidetracking, and studied the criteria for their applicability.

By applying cluster analysis to the selected wells, three clusters were formed, each including three wells, united by the geological properties of their reservoir rocks and the filtration–capacitive properties of the oil. After this, the optimal technologies were selected for two clusters—hydraulic fracturing and sidetracking. For the remaining cluster, it was decided that it would be inappropriate to use hydraulic fracturing and sidetracking for its wells. Thus, a promising solution to the technical problem was proposed, which can bring a significant technical and economic effect in the development of the oil field under consideration.

Further, analytical formulas and mathematical modeling were used to calculate the technological effects for each of the selected wells: the average oil flow rate for the three wells where sidetracking was performed was 25.53 t per day, and that for the wells where hydraulic fracturing was performed was 17.01 t per day. The accumulated oil production, recovered due to the application of the technologies, from six wells for the first 7 years after the operation is estimated at 306.92 thousand tons of oil. Due to the achieved technological effect, the economic efficiency of development of the studied oil field will increase due to the proceeds from the sale of the additional extracted oil.