A Methodology for Forecasting the KPIs of a Region’s Development: Case of the Russian Arctic

Abstract

Forecasting the development of regions is one of the most challenging tasks of modern economics. The quality of any forecast is determined by the methodology used. Accordingly, criticism of existing forecasts is largely connected to their methodological approaches. In this paper, a multi-level approach to forecasting the development of the region is proposed, starting with the definition of the key performance indicators and ending with the assessment of various scenarios. The study was conducted on the example of the Russian Arctic, divided into three technological zones, with three scenarios of the development for each (negative, base, positive). The application of the proposed methodology allowed for modeling the development of the region until 2035. The results show that the Russian Arctic has a huge difference in the achievability of different goals, e.g., 98% of the electricity supply targets are achievable in a baseline scenario, while only 52% are achievable in a set of “navigation” targets. The proposed methodology can be useful for diving into the details of regional forecasts, such as the impact of key companies in a region or the influence of international politics.

1. Introduction

From a sustainable development perspective, it is critical for the government to predict the future demand for the sectoral raw materials and the level of the following industrial development. In addition to demand, industries are affected by external challenges, risks in all spheres of life, as well as by modern trends and new technologies. The rapid technological growth in the middle of the period under review can seriously affect the industry. Impacts may be both positive (for example, am efficiency increase of hydrocarbon production) and negative (for example, the obsolescence of the other equipment and the financial losses of the state and companies).

Long-term forecasts should be relied on when building strategies for the country’s development with yet another aim of preserving sovereignty, which is crucial in conditions of economic and political instability. In this way, any country can achieve human, technological, mineral, and energy self-sufficiency as well as reorganize its trade relations [1,2]. To reach these goals, it is essential for states to recognize their strengths to intelligently manage unique capabilities and innovative technologies. During building state programs and forecasts, it is necessary to consider the territories of innovative development, which have a certain specificity and attractiveness [3].

The Arctic Zone of the Russian Federation (AZRF) is not just a region with a unique climate and northern nature. It is the world’s largest special economic zone, with a specific tax and administrative system. At the same time, related to the Arctic Zone, the Northern Sea Route (NSR) is the shortest route from Europe to the Asia-Pacific countries and a real alternative to the route through the Suez Canal on the near-term planning horizon. For the whole world, the use of this transport corridor may significantly reduce the cost of cargo freight. In this scenario, the AZRF would become a major center of international trade, as well as a platform for research.

For the Russian Federation, the Arctic Zone is recognized as the most important vector of development at the state level. This is confirmed by dozens of investment projects and regional development programs. In addition, the future of AZRF is actively discussed in the international scientific community: scientific centers are established, large annual conferences are held, and young scientists are invited to develop modern plans.

The Arctic is a complex region with a heterogeneous structure, so there are still many questions that need to be addressed. It is a territory of contrasts, where mineral and raw material potential and special environmental vulnerability are combined, and one of the main sea transportation arteries borders poorly developed continental infrastructure. Given the combination of these factors, Arctic infrastructure requires a comprehensive approach to its development.

The attractiveness of the Arctic is reinforced by a host of external (global) and internal challenges. These include, in particular, general population growth, an unstable geopolitical situation in the face of market reorientation, increased consumption in the face of resource depletion, and climate change. A number of Sustainable Development Goals (SDGs)—8 (decent work and economic growth), 11 (sustainable cities and human settlements), 12 (responsible consumption and production), and 13 (combating climate change)—have been proposed as responses to these global challenges [4,5]. The AZRF can become a testing ground for sustainable technologies and development approaches corresponding to the SDGs.

The NSR is one of the main vectors of the development of the AZRF, which requires the implementation of extraction and processing projects (especially those related to liquefied natural gas (LNG)), ensuring safe navigation for export to the Asia-Pacific Region (APR), and the creation of autonomous energy supply systems [6,7]. The amount of funding for the implementation of government plans in the Arctic reaches up to USD 19.34 billion [8,9]. Earlier, the goal was set to achieve a cargo traffic of 80 mln tons by the current year while ensuring year-round navigation in the eastern section of the NSR [10,11]. This figure should double by 2035.

To achieve future targets, it is necessary to assess the current situation of the territorial clusters of the AZRF and the NSR and select key performance indicators (KPIs) that can characterize their development. The basis for forecasting KPIs development is the primary consideration of external and internal conditions [12]. They directly or indirectly influence quantitative risk assessments and set the vector of the territory’s development.

The main goal of the article was to form a forecasting model that takes into account the risks associated with the development of the territorial cluster under consideration as well as the impact of aspects of technological progress and regional policy on the forecast values.

The objectives included

-

the evaluation of existing forecasting approaches, highlighting their strengths and weaknesses to create a hybrid method;

-

the search for the KPI of the regional development;

-

the assessment of the degree of related factors’ influence;

-

the modelling under consideration of the aforesaid factors.

This paper is further structured as follows: Section 2 presents the literature review. Section 3 describes the methodology used. Section 4 summarizes the empirical results. Section 5 is devoted to conclusions.

2. Literature Overview

2.1. Methods of Regional Forecasting

The authors considered various approaches to regional forecasting, taking into account their disadvantages, in order to form a better complex model. In a paper [13], the socio-economic development model takes into account external parameters (macroeconomic indicators) and the logical interrelation of indicators based on the correlation and regression analysis of development scenarios and balance ratios. One of the disadvantages of the method is that the result is programmed by the properties of the model, which the authors can explicitly or implicitly introduce into the structure itself, predetermining its result.

A similarly targeted work [14] examined urban economic development at the cadastral site scale based on the modeling of sectoral land use dynamics and the S-curve model. The resulting economic indicators of growth were presented for each sector and assigned at the cadastral scale based on the sectoral land use type, land area, and location attributes. A disadvantage of the approach is its low scalability and it building on prior experience. It is impossible to both introduce new technologies into the model and account for current events and risks.

The issues of the approbation of the model data were also considered. The authors in [15] presented an approach in which, in order to level the problem of long sequences validation, the TCN (Temporal Convolutional Network) model was presented, which implements the fixation of dependencies over extensive time intervals and offers efficient model training. However, when significant dynamic changes in time occur within long data sequences, the accuracy of the described model decreases.

In turn, the MAcroeconometric, Social, Sectoral, Territorial fifth generation model (MASST) is represented in a paper [16]. The model allows for producing numerical forecasts of global worlds’ regions’ development considering various factors. The forecasts are produced on the basis of linear equations, considering target indicators and the speed of the indicators change. The disadvantage of this model is the necessity to change its structure significantly to include external shock-influences (like pandemics and war conflicts) into the forecasts.

The expert forecasting model or hybrid intellectual system is presented in an article [17] and taken into account by the authors. The main goal of the system is predicting the values of socio-economic development indicators. The system has been tested in predicting more than 600 parameters and showed high efficiency. In the base of the study, a neural network principle is used, which requires high computational power and time. The perceptrons are individually created for each of the socio-economic parameters to be predicted, and this is the main disadvantage of the proposed system.

A multilayer neural network-based model for regional macroeconomic forecasting was invented by [18]. The verification of the model has been provided with the help of the Chinese regional economics GRP values. The model itself provides the calculations using the down-up principle, starting from the lower information and reaching the complex unified indicators of economic growth separately for the three layers of industries presented in the region. The model is highly suitable for macroeconomic studies, but it is not informative for specific region investigations.

Another work that proves the effectiveness of neural network use in forecasts is [19]. The authors provide a comparative study to prove the priority of neural networks over other linear statistical prediction methods. However, the analysis does not provide any conclusions about expert models involving people in the process of forecasting. The proposed neural model is rather easy and does not really include many parameters and factors in the process. It also does not result in more than one indicator.

A mathematical forecast model is presented in a study [20]. Generally, the model includes three stages following each other, which are the volume of investment forecast, production factors forecast, and gross regional project (GRP) forecast. All three forecasts are based on statistics analysis methods, allowing for finding out the trends expected on short- and long-term periods. The disadvantage of the proposed model is the limitation of the outcome data. The resulting indicator is only GRP, which does not allow for creating full-fledged strategies for the regional development based on the forecast.

An economy-cycle system is represented in a study [21]. The system allows for forecasting the crucial phenomena that could possibly happen in the economy lifecycle of the country. The verification of the model was held on the example of the industry production parameters and showed high efficiency. However, the forecasting model includes many calculation steps and requires them for every predicted indicator. The system requires high computational power.

A paper [22] represents a special exercise for an expert group which aimed to create a strategy of the energy sector development of the region. The exercise methodology is very much similar to the model described in this study. It also includes the PESTEL and SWOT methods, first to identify and classify the main driving forces and then to create the strategy guideline. This approach does not include technology cases analysis and their implementation, which may improve the created strategies.

The provided literature review shows that most of the presented systems are verified on the real datasets and show their efficiency. However, most of the models are difficult to reproduce, as they use neural network algorithms, require precise tuning for every specific case-study, and cannot be unified for predicting different indicators at once. Some models also do not consider further strategy planning, usually following the forecasting process.

There is a need for a regional forecasting model that is rather easy to be reproduced, includes current tendencies, trends, and risks of the regional development, and also provides specific scenarios depending on the induced governmental measures.

2.2. Overview of the AZRF Industries

The Arctic is one of the most challenging regions to forecast. The development of various industrial sectors in the AZRF itself requires investments from other regions of the country. The authors analyzed the readiness of the Russian Federation industries to meet the anticipated needs of the AZRF. The industries under consideration are the machine-building, metallurgy, timber, and chemical industries [23,24]. The main centers of sectoral development and suggested methods of transportation are presented in Figure 1.

In the Russian context, significant attention should be paid to mechanical engineering. There are successful domestic projects in this sector, such as the Arctic Cascade technology in the 4th line of Yamal LNG. Results were also obtained in the field of submersible oil production, pipeline valves, and pumping and drilling equipment. Nevertheless, the share of Russian equipment for geodesy, cartography, seismic exploration, associated petroleum gas (APG) utilization, and process automation is still extremely small. Also, units for LNG projects and hydraulic fracturing are critically dependent on foreign supplies. Only 0.9 out of 30 mln tons of the LNG plant capacity installed in Russia today uses domestic equipment [25]. The withdrawal of a number of companies from the joint projects of PJSC Gazprom and PJSC Novatek makes it difficult to commission LNG facilities for 140 mln tons per year by 2035 [26].

In the energy sector, the restriction of access to combined cycle gas turbines (CCGT) and gas turbine units (GTU) has become tangible. The withdrawal of Siemens, General Electric, and Alstom may have significant consequences in the long term, so far solved through parallel imports.

Metal-intensive infrastructure projects in the AZRF depend on the level of development of the metallurgical industry. Imports of certain categories of products are already reduced to zero. Such industry leaders as PJSC MMK and PGSC Severstal are implementing an import substitution program. However, the share of domestic products in some categories is insignificant and will reach the target values only this year.

The timber industry is a specialization of the European North of Russia (Arkhangelsk region and Komi Republic) and is sufficiently developed. The availability of a raw material base is the main condition for the distribution of enterprises, contributing to the implementation of projects in the AZRF.

In the chemical industry, the largest enterprises are located in the Volgo-Vyatsky, Central and North-Western economic regions. Despite the development of the industry, many of the products required for the AZRF are imported.

If the industry potential of the Arctic is to be implemented, it is also important to take into account the domestic markets for the products. The location of processing centers and supply routes from the AZRF is shown in Figure 2.

According to the forecasts presented in

Table 1. Planned volumes of oil and coal production in the AZRF until 2035.

Region (Resource)	Production Volume in 2021, mln Tons [28]	Production Volume in 2025, mln Tons [29]	Production Volume in 2030, mln Tons	Production Volume in 2035, mln Tons
YaNAO (oil)	35.983	60.803	84.986	85.599
NAO (oil)	11.507	22.534	27.794	23.308
Krasnoyarsk Krai (oil)	15.641	23.083	19.961	17.648
Komi Republic (oil)	13.4	14.57	–	–
Krasnoyarsk Krai (coal)	0.4	5	10	10
Komi Republic (coal)	8.85	9.2	10.2	6.7
Chukotka A. Okrug (coal)	0.85	2	2	2

, the oil and gas and coal industries are projected to develop the most in the AZRF. For 2021, coal production in the AZRF amounted to 2% of the value for the country [27].

In its turn, gas production in 2021 in the AZRF amounted to 606 billion m³ (87.4% of the RF production) [30]. About 95% of these volumes fell on the Yamalo-Nenets Autonomous Okrug, which is presented in

Table 2. Planned gas production volumes in the AZRF until 2035.

Region	Production Volume in 2021, bln m3 [31]	Production Volume in 2025, bln m3 [29]	Production Volume in 2030, bln m3	Production Volume in 2035, bln m3
NAO	0.18	0.407	4.46	6.30
YaNAO	598.67	672.43	663.83	597.5
Krasnoyarsk Krai	7.11	9.62	7.28	6.92

.

The Arctic gas industry itself is linked to LNG production. This is the strategic direction of PJSC Novatek, which is implementing the Arctic LNG 2 project. The production volumes in the territory of the AZRF are forecasted to reach 91 mln tons by 2035 [32].

The competitive advantage of most Arctic industries is their proximity to the NSR. This opens up opportunities for export activities with Asia-Pacific countries. Crude oil is delivered to the consumer by oil pipelines, and after that, in tankers accompanied by icebreakers.

Most of the oil and gas refineries in Russia are currently in the process of modernization. The slow development of the industry is due to taxation and high import dependence. Most of the capacities are concentrated in western and central Russia. The readiness for oil refining is low.

The LNG industry development is more dynamic, with a trend of 83 t/h installed capacity growth for 2021. The industry is ready to supply products from AZRF, but there may be a shortage of processing capacity in the country.

Ore mining and processing in the Arctic has ambiguous potential. Despite 90% of imports of rare earth metals (REMs) being from abroad, the further development of this area is possible under the guidance of Rosatom. Half of the existing REMs concentrate production plants are located in the AZRF. The obtained concentrate is sent for processing to the Solikamsk Magnesium Plant. The Tomtor deposit (Sakha Republic) has special potential, with an annual design capacity of 160 thousand tons of ferroniobium [33]. Krasnoyarsk platinum deposits of PJSC MMC Nornickel provide up to 96.1% of production in the RF.

Although Rosatom is implementing the REMs development program, according to which the Russian Federation will be able to fully provide itself with the resource by 2030 (in the amount of 7.5 thousand tons per year [34,35], the number of processing facilities is currently insufficient. The Solikamsk magnesium plant is the only one in the country with four concentrate production plants.

The presented complex analysis allowed the authors to divide the territory of the AZRF into three enlarged territorial clusters–aqua-territorial-production clusters (ATPCs): western, central, and eastern [36]. In Figure 1 and Figure 2, they are highlighted in color in accordance with the specifics of production, the geological and geographical conditions, the geopolitical position, and the regional economic situation. When implementing strategies, it is necessary to take into account the whole complex of these specific features of ATPCs. A similar division can be achieved when analyzing other territories in order to build a prognosis, taking into account the technological impact.

3. Materials and Methods

The approach developed and described further by the authors is based on the change in scenario values under the influence of zonal (cluster) risks. It is assumed that the risks will be influenced by the leading companies, technological cases (for example, in [37,38,39]), and proposed measures for a given time period.

When building the forecast mathematical model, the authors decided to take the quantitative values of zonally distributed risks (X_n) as the starting point of internal and external conditions. They characterize the current state of the regions and, in the aggregate, affect the scenario values of KPIs (Y_n) at positive, negative, and baseline forecasts. The degree of this influence depends on the factors of the external environment (the policies of leading companies, technologies, and proposed integrated measures).

The mentioned influence was reflected by the change in quantitative values throughout the “path” of each risk from the initial value of X_n to the value of X_n‴. In turn, all X_n‴ influenced the scenario values of Y_n. The basis of the whole research is schematized in Appendix A in Figure A1, particularly in Figure 3.

Eight metrics of KPIs Y_n were considered as predicted scenario values for two groups: NSR development and the development of energy infrastructure of the NSR and AZRF. They included the total volume of transportation along the NSR (mln tons/year); the average ratio of icebreakers’ number and capacity (pcs./MW); the NSR navigation period (days/year); CO₂-eq. emissions from non-LNG ships [40] (mln tons CO₂-eq./year); the installed capacity of power plants in the AZRF (MW); the specific fuel consumption for electricity supply in the AZRF (g c.f./kW∙h); the specific fuel consumption for heat supply in the AZRF (g c.f./Gcal); and the total production in the AZRF (t c.f./year).

The first step in building the mathematical model was the transition from qualitative risks to their quantitative expression. A total of 132 risks were analyzed by the authors, but their number was later reduced to 36. The expert judgements were obtained with use of the Delphi method. The risks selected for further work according to the results of the first monthly survey (among 32 experts on Arctic topics) turned out to be the most important in terms of the degree of their influence, with the highest probability of occurrence. Their numerical expressions X_n were found for these risks.

Further, it was necessary to assess the influence of the companies’ policies on the values of risks. A ranking of 12 leaders by their presence in the AZRF was made by assessing economic performance, compliance with ESG standards, and localization. As a result, the impact of the companies’ policies was reflected in the form of curves of changes in the value of risks X_n′.

The previous stage indirectly covered the impact of technological development on risk reduction. They were fully taken into account when considering the technological cases of the companies. Fifteen blocks of technologies, combining 218 cases, take into account the trading potential of a much larger number of suppliers from the Russian Federation and foreign countries. To select a list of priority technology blocks and relevant cases, an additional comparative analysis was carried out using the hierarchy method in three ATPCs. After that, the impact of technological solutions for ATPC on risks was assessed and X_n″ values were obtained.

The degree of influence of the companies’ cases was determined through the weighting coefficients obtained by the results of the second monthly survey among a wider range of respondents (50 representatives of fuel and energy companies, scientific centers, and educational organizations). The expert influence coefficients were obtained in a similar manner to how they were obtained in the first survey, with the use of the Delphi method. The curves obtained earlier changed in proportion to the degree of influence, keeping the same trends as those obtained during the first transformation.

The last step in the transformation of risks was the impact of regulatory measures on them. Measures for compensating, strengthening, implementing, and capturing the existing aspects of development were formulated according to the CARD (compensators, amplifiers, realizators, and deflectors) methodology.

At the last stage of risk management, the overall impact of all PESTEL (political, economic, social, technological, environmental, and legislative) risks on each of the forecast values was found under three scenarios.

The authors’ approach is a combination of existing and widely used methods. PESTEL and CARD classifications were used to identify risks and responses, respectively. PESTEL has been improved by adding the dynamics of risk change over time. The years of implementation of measures and the extent of their realization were also integrated with the CARD approach. The Delphi method is used to systematize the collection of expert opinions in two surveys aimed at ranking risks and determining the extent to which they are influenced by a number of factors. These include CARD measures and companies’ cases. To obtain objective results, the obtained values of the coefficients in both cases were averaged. In the process of the selection of multidirectional technological innovations, the hierarchy method was used to compare them in pairs. In the standard case, it is used for the subjective comparison of solutions to strategic problems. When considering technological approaches, scores were assigned on the basis of numerical indicators of their efficiency.

3.1. Identification and Quantification of Risks

Strategic planning implies taking into account all development-related issues when making management decisions. Integrated development involves a number of risks related to its political and economic aspects, social problems in the regions, and the level of their technological endowment, as well as the environmental and legislative peculiarities of ATPCs. The identification was carried out according to the principle of PESTEL analysis in order to take into account all of the above. At the first stage, a thesis list of risks for each of the ATPCs was formed (presented in

Table A1. List of qualitative risks in the western ATPC of the AZRF.

P	E	S	T	E	L
Weak trade advantages compared to EU countries and central regions (in terms of energy prices and product quality)	Unfavorable financing terms (due to tighter credit conditions, leasing, and microfinance lagging behind)	Unemployment (job losses due to falling production and the depletion of major fields)	Territorial constraints (for the construction of port and mining infrastructure)	Coastal abrasion	Establishment of new expenditure commitments
Threat to national safety (including threat to production facilities)	Volatility of prices and demand for energy resources in the domestic market (due to the growth of tariffs for natural monopolies)	Differentiation of the socio-economic development of municipalities	Depletion of an easily accessible mineral resource base (Cu-Ni, apatite and iron ores, etc.)	Increase in waste volumes	Lack of specific technical legislation (e.g., in the use of digital financial assets)
Suspension of membership in international projects	Changing priorities of economic development (shift from extraction of fossil resources)	Population aging	Low balance of the transportation network (railway approaches to the ports of Murmansk and Kandalaksha; lack of bridges)	Impact of objects of accumulated harm	Tightening of environmental quality standards (entailing an increase in related expenditure commitments)
Slowdown in the implementation of national projects and federal programs	Logistics costs (due to the dispersion of settlements, their inaccessibility, and the deterioration of transport infrastructure, especially from the position of railways)	Shortage of qualified personnel	Insufficient volume and quality of geological exploration works	Weak ecosystem resilience (fire hazard)	Lack of measures for attracting funds from parent companies to subsidiaries (in terms of tax payments)
Trading isolation	Falling investment attractiveness and activity (reduced capital investments in the fuel and energy complex)	Complicated access to social services	Low development of the sphere of construction materials production	Growing man-made and anthropogenic impact (including tourism)	Obstruction of trade policy (with unfriendly states)
Technological isolation	Foreign companies leaving	Migration outflow of population (socio-economic)	Disproportion in the location of the main generating capacities	Accidents at production sites	Changes in legislation in the field of subsoil use licensing
Transportation and logistical isolation	Conflict between recreational and productive functions of regions (difficulties in prioritizing development)	Deterioration of the quality of healthcare	Low commercialization of innovations (in digital and manufacturing technologies)	Soil thawing	Changes in legal requirements in the field of the design, construction, and operation of facilities
Global economic crisis (decline in world trade and production)	Financial losses due to state tariff regulation of gas prices	Risk of epidemics	Wear and tear of energy infrastructure	Accidents on transportation routes (NSR and pipelines)	Tightening of fuel requirements for transportation vessels

,

Table A2. List of qualitative risks in the central ATPC of the AZRF.

P	E	S	T	E	L
Transportation and logistical isolation	Volatility of prices for energy resources	Migration outflow of the population (ecological, socio-economic)	Depreciation of energy infrastructure	Growing technogenic and anthropogenic impact by the mining sector (as a consequence, a reduction in agricultural areas)	Lack of specific technical legislation
Trading isolation	Falling investment attractiveness and activity (as a consequence, increased implementation time)	Low level of qualification (aggravated by the lack of educational centers)	Underdeveloped energy connections	Secondary contamination of drinking water in water supply networks	Lack of measures for attracting funds from parent companies to subsidiaries (in terms of tax payments)
Technological isolation	Increase in expenditures for the repair and maintenance of facilities	Population aging (hence, an increase in the demographic burden on the able-bodied population)	Gas supply disruptions (e.g., in case of damage to the Longyugan-Salekhard gas pipeline)	Long periods of unfavorable meteorological conditions (poor dispersion of pollutants)	Establishment of new expenditure commitments
Suspension of membership in international projects	Unfavorable financing terms (due to tighter credit conditions, leasing, and microfinance lagging behind)	Complicated social and labor conditions (difficult adaptation of labor migrants and wage debts)	Disruption of heat supply (as climatic conditions affect the stability of the systems)	Growing technogenic and anthropogenic impact by the metallurgical sector	Tightening of environmental quality standards
Unequal socio-economic support to the centers (mostly applies to the Siberian Federal District)	Export-oriented structure of the economy (imbalance of imports and exports)	Risk of epidemics (due to climate change)	Lack of technology clusters (in the form of R&D centers)	Accidents on transportation routes	Lack of consideration of the specifics of remote regions
Global economic crisis	Underpayment of taxes	Hidden unemployment	Low commercialization of innovations (in digital and manufacturing technologies)	Accidents at production sites (especially in LNG production)	Obstruction of trade policy
Slowdown in the implementation of national projects and federal programs	Logistics costs (due to the dispersion of settlements, their inaccessibility, and the deterioration of transportation infrastructure)	Deterioration of the quality of healthcare (low availability of medicines)	Low development of the sphere of construction materials production	Growth of industrial waste volumes	Establishment of conditions for the passage of ships through the NSR

and

Table A3. List of qualitative risks in the eastern ATPC of the AZRF.

P	E	S	T	E	L
Global economic crisis	Falling investment attractiveness and activity (as a consequence, increased implementation time)	Low level of qualification (reflected in the absence of educational centers)	Accelerated transition to combined power supply sources	Coastal abrasion	Lack of specific technical legislation
Transportation and logistical isolation	Volatility of prices for energy resources	Unemployment	Depreciation of energy infrastructure	Accidents at production sites	Tightening of environmental quality standards
Trading isolation	Supply disruptions during the inter-navigation period (the risk relates to the importation of agricultural products)	Migration outflow of the population (ecological, socio-economic)	Complicated construction of transportation infrastructure	Accidents at production sites (especially in LNG production)	Lack of consideration of the specifics of remote regions
Technological isolation	Planned unprofitability of the region (lack of funds for maintaining energy and utility infrastructure)	Risk of epidemics (due to climate change)	Low commercialization of innovations (in digital and manufacturing technologies)	Permafrost thawing (release of bound carbon, reduction in soil bearing capacity)	Establishment of new expenditure commitments
Threat to the safety of passage through the Bering Strait	Dependence of the county’s economy on precious metals mining	Lack of comfortable housing conditions (high deterioration of the housing stock, lack of quality domestic water supply)	Lack of technology clusters (in the form of R&D centers)	Destruction of natural geochemical barriers	Obstruction of trade policy
Suspension of membership in international projects	Increase in expenditures for the repair and maintenance of facilities	Complicated social and labor conditions (complexity of adaptation of labor migrants)	Disproportion in the location of the main generating capacities	Weak ecosystem resilience	Lack of measures for attracting funds from parent companies to subsidiaries
Slowdown in the implementation of national projects and federal programs	Unfavorable financing terms (due to tighter credit conditions, leasing, and microfinance lagging behind)	Deterioration of the quality of healthcare (low availability of medicines)	Low development of the sphere of construction materials production	Lack of large-scale emergency strongholds	Establishment of conditions for the passage of ships through the NSR

in Appendix B). Priority was given to the phenomena mentioned in the current programs of the socio-economic development of the regions.

In the western ATPC, much is determined by the specifics of the areas bordering neighboring countries. The influence of the country’s isolation and the manifestation of internal peculiarities—a low level of infrastructure development and a vulnerable position of the environment under an increasing impact on it—are noticeable. At the same time, ATPC is characterized by disproportion in the location of the main generating capacities.

In the central ATPC, greater emphasis is placed on the wear and tear of infrastructure (especially electricity and heat), with a simultaneous increase in repair and maintenance costs, as well as on the risks associated with extensive technological development.

The eastern ATPC is characterized by risks associated with limited navigation, which affects the standard of living of the population and the rate of migration outflow [41,42]. The complicated construction of transportation infrastructure under the condition of soil thawing aggravates the current problems. The wear and tear of energy facilities combined with the forced transition to combined sources of energy supply affect the continuity of the energy supply of the entire ATPC.

The approach presented by the authors implies the use of quantitative values of risks. With the appearance of numerical expressions, it becomes possible to trace the influence of external conditions on them and, subsequently, on the total values. Some risks are to be identified in each category (P, E, S, T, E, L) for each ATPC as the most important. They are the most likely to occur and have the strongest impact within their territorial area, as assessed by experts in the field of Arctic research. While the units of measurement and the risks themselves could be repeated between the regions, their quantitative values differed from region to region.

Quantitative values for risk description were taken from official information sources (Federal State Statistics Service; official websites of the Ministry of Energy of the Russian Federation, Russian Railways, Ministry of Emergency Situations of the Russian Federation, Ministry of Natural Resources, and regional governments). All of them characterize the current situation in ATPCs.

3.2. Influence of Companies on the Quantitative Value of Risks

When assessing the impact of the companies’ policies on the quantitative values of risks, it was required to reduce their list to those really significant in the AZRF. For this purpose, a ranking was compiled. The ranking was carried out in terms of the economic and environmental sustainability of organizations, as well as their direct presence in the AZRF. On this basis, three criteria were used for the aggregate assessment: a specific economic indicator, compliance with ESG standards, and the degree of localization in the Arctic zone. For each of them, the maximum value was 1 and the minimum value was 0.

The economic indicator consisted of three parts: ROE (return on equity), CapEx/EBITDA (ratio of capital expenditures to profit before all expenses), and GAGR NI (average annual net profit growth over the last 5 years).

For the correct comparison of companies by ROE and CapEx/EBITDA, the average and maximum values in the industry among the largest public companies by capitalization were calculated. According to the results, the industry average value of ROE and CapEx/EBITDA corresponded to the coefficient of 0.5, and the maximum corresponded to 1. Intermediate values were calculated by interpolation.

When using the GAGR NI criterion, a coefficient of 0.33 corresponded to the value of 13% p.a. and 0.66 corresponded to the value of 26% p.a. The maximum value among the companies considered was 1. The value of 13% was chosen based on the doubled official average inflation rate in the Russian Federation over the last 5 years. Intermediate values were calculated by interpolation.

ESG compliance was assessed based on the calculations of RAEX and NRA ranking agencies.

It was not correct to compare the degree of localization in AZRF in a single way for all companies due to the different focus of their activities. On this basis, the calculation was carried out according to Formulas (1)–(5):

$$\mathrm{For} \mathrm{oil} \mathrm{and} \mathrm{gas} companies: k={\displaystyle \frac{\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{s} \mathrm{i}\mathrm{n} \mathrm{W}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{n} \mathrm{S}\mathrm{i}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{a}}{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{s}}},$$

(1)

$$\mathrm{For} \mathrm{PJSC} \mathrm{GMK} Nornickel: k={\displaystyle \frac{\mathrm{o}\mathrm{r}\mathrm{e} \mathrm{e}\mathrm{x}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{s} \mathrm{i}\mathrm{n} \mathrm{K}\mathrm{o}\mathrm{l}\mathrm{a} \mathrm{a}\mathrm{n}\mathrm{d} \mathrm{N}\mathrm{o}\mathrm{r}\mathrm{i}\mathrm{l}\mathrm{s}\mathrm{k} \mathrm{d}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s}}{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{o}\mathrm{r}\mathrm{e} \mathrm{e}\mathrm{x}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{s}}},$$

(2)

$$\mathrm{For} \mathrm{PJSC} Transneft: k={\displaystyle \frac{\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{r}\mathrm{u}\mathrm{n}\mathrm{k} \mathrm{p}\mathrm{i}\mathrm{p}\mathrm{e}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{s} \mathrm{i}\mathrm{n} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{A}\mathrm{r}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{c}}{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{r}\mathrm{u}\mathrm{n}\mathrm{k} \mathrm{p}\mathrm{i}\mathrm{p}\mathrm{e}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{s}}},$$

(3)

$$\mathrm{For} \mathrm{UC} Rusal: k={\displaystyle \frac{\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{u}\mathrm{m} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{A}\mathrm{r}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{c} \mathrm{s}\mathrm{m}\mathrm{e}\mathrm{l}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{s}}{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{a}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{u}\mathrm{m} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{s}}},$$

(4)

$$\mathrm{For} \mathrm{PJSC} Sovcomflot: k={\displaystyle \frac{\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{A}\mathrm{r}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{c} \mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{s} \mathrm{i}\mathrm{n}\mathrm{v}\mathrm{o}\mathrm{l}\mathrm{v}\mathrm{e}\mathrm{d}}{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{s} \mathrm{i}\mathrm{n}\mathrm{v}\mathrm{o}\mathrm{l}\mathrm{v}\mathrm{e}\mathrm{d}}}.$$

(5)

Based on the business plans and strategies, the model first set the maximum possible exponential or linear changes in risks and then built the corresponding matrices of their values X_n′.

3.3. Influence of Technological Cases on the Quantitative Value of Risks

The basis for the functioning of the previously mentioned companies is their technological security and energy safety. In some cases, the needs are covered by their own developments, but often supplies from other manufacturers are necessary. For this reason, organizations from central Russia and abroad were considered when evaluating the cases.

The technological cases of companies affect not only the optimization of production processes but also the areas of life support and environmental protection, and they stimulate the development of legislation. In order to take into account all interrelationships, it was necessary to carry out a comparative analysis of cases in technological blocks by the hierarchy method using PESTEL risks as “criteria”.

The technological blocks under consideration included exploration, extraction, processing, utilization, aircraft construction, river shipbuilding, marine shipbuilding, navigation and logistics, new materials technologies, construction, power generation, power transmission, power consumption, heat and power engineering, and non-traditional renewable energy sources.

The main task in the comparative analysis using the hierarchy method was to identify the cases in their block of technologies that have the greatest impact on overcoming risks for each ATPC. In their turn, the risks were preliminarily assessed in terms of the degree of their impact on the development of the energy infrastructure of the AZRF and the NSR as a whole. Matrix pairwise comparisons were carried out in accordance with Figure 4.

Initially for the “criteria” and further for the cases, the assessment was carried out on a qualitative scale with conversion into points, where the options are equivalent—1; the option is slightly weightier (less weighty) than another—3 (1/3); the option is weightier (less weighty) than another—5 (1/5); or the option is significantly weightier (less weighty) than another—7 (1/7). As a result, the authors built matrices of pairwise comparisons on the example of

Table 3. Matrix of pairwise comparisons of risks.

Criteria	P	E	S	T	E	L	Weight, %
P	1.00	3.00	3.00	3.00	0.20	1.00	21.46
E	0.33	1.00	3.00	3.00	1.00	3.00	19.93
S	0.33	0.33	1.00	1.00	0.20	5.00	10.22
T	0.33	0.33	1.00	1.00	1.00	5.00	13.79
E	5.00	1.00	5.00	1.00	1.00	3.00	28.56
L	1.00	0.33	0.20	0.20	0.33	1.00	6.04

by “criteria” (risks), but already for the cases.

In the technology blocks, all related cases were compared with each other in terms of their impact on each of the PESTEL risks (one comparison table per risk). Using the results of the identification of significant cases by technology blocks in ATPCs, a second survey was conducted among a wide range of experts from industry and educational institutions. The results were processed by means of finding median values of assessments of the degree of influence of the companies’ cases on risks (from 0—no influence to 10—maximum influence) and obtaining a matrix of X_n″ values.

3.4. Impact of the Proposed Measures on the Quantitative Value of Risks

The measures compensators, amplifiers, regulators, and reflectors are distributed unevenly across the programs (involving combinations of measures), as the choice of measures depends on the specifics of the group of risks, which the program is aimed at eliminating to a greater extent.

The degree of impact of measures on risks was assessed by experts based on the results of the second survey. Similar to the previous stage, the results were processed by finding the median values of these influences (from 0—no influence to 10—maximum influence). Further on, they were taken into account in finding the matrix of X_n‴ values.

While the considered blocks of technologies and the influence of company policies did not vary by scenarios, the degree of implementation of the programs with CARD measures and the years of their introduction did. Thus, the risk values X_n‴ were obtained within each ATPC for the three scenarios with their own value.

Each measure according to the results of the survey had a common degree of influence on the risk X_n″ (a), different scenario degrees of implementation (b), and its periods (c—year of beginning and d—year of ending). The maximum impact of a measure on the risk for its adjustment was calculated by multiplying the value of X_n″ in 2035 obtained in the previous steps by the coefficients a and b. The impact of measures was taken into account through graphs, which were plotted according to condition (6) (where x is the year):

$$\left\{\begin{array}{c}\mathrm{when} 2022\le x<c, y=0 \\ \mathrm{when} c\le x\le d, y=k·x, \mathrm{where} k \mathrm{is} \mathrm{the} \mathrm{slope} \mathrm{factor}\\ \mathrm{when} d<x, y=a \end{array}\right.$$

(6)

X_n″ either increased or decreased under the influence of the measures (depending on the direction of the positive effect). At this stage, three to five curves were obtained for each risk for each of the three scenarios. Further, the total impact of measures on the individual risk was found. Thus, new curves X_n‴ were obtained.

The moduli of risk changes were reported in percentages relative to the 2022 level, after which the authors obtained the total relative risk change for each of the items in the PESTEL analysis for the three ATPCs by adding the changes in “neighboring” risks (two political risks within one ATPC, etc.).

The weighted average influence of risks of one direction on KPIs was found for the transition from risks to changes in forecast values. The share contribution of the influence of each direction on the final change in Y_n was estimated based on the second survey, according to the condition of which the maximum total influence of risks should not exceed 100%. Thus, the influence of each of the six generalized PESTEL risks on all eight forecast values for the three scenarios was obtained.

To link X_n‴ and Y_n, the dynamics of Y_n from 2015 to 2022 were taken into account, as well as their baseline approximation up to 2035. The typical increment of values at this period was multiplied by the total impact and summarized with the baseline approximation. The year of 2022 was taken as the base year for the normalization of values and the construction of graphs by the method of relative comparison.

Thus, it became possible to construct three graphs reflecting the trend of forecast values from 2015 to 2035 and obtain their direct values.

4. Results and Discussion

The reduction in the list of the initial 132 risks from Table A1, Table A2 and Table A3 in Appendix B was carried out by means of estimation by an expert group of their probability of occurrence and degree of influence on a horizon until 2035 (with use of the Delfi method).

The authors identified two risks in each category (P, E, S, T, E, L) for each ATPC (36 in total) as the most important. At the same time, their quantitative values sometimes differed from region to region. For example, the risk “decrease in investment attractiveness and activity”, expressed in terms of its growth rate from 2016 to 2020, amounted to 2% in the central ATPC, while in the eastern ATPC, it was 12% (see

Table A4. Quantitative values of principal risks in 2023.

Risk/ATPC	Western ATPC		Central ATPC		Eastern ATPC
P	Threat to national safety	Trading isolation	Transportation and logistical isolation	Technological isolation	Transportation and logistical isolation	Trading isolation
	Number of attacks on refineries and other industrial facilities of the fuel and energy complex in 2022: 11	Export and import growth rate: 1.41%	Number of railroad tracks from the “big land” to the region: 4	The level of availability of industry technology: 70%	Number of railroad tracks from the “big land” to the region: 1	Export and import growth rate: 1.5%
E	Unfavorable financing terms	Logistics costs	Falling investment attractiveness and activity	Increase in expenditures for the repair and maintenance of facilities	Falling investment attractiveness and activity	Supply disruptions during the inter-navigation period
	Insurance contribution rate for AZRF residents: 7.5%	Freight transportation tariff index growth rate: 33.8%	Growth rate of investment activity: 2%	Share of emergency housing stock: 10%	Rate of decrease in investment activity: 12%	Aviation mobility: 0.58 trips per 1 person/year
S	Shortage of qualified personnel	Migration outflow the of population	Migration outflow of the population	Low level of qualification	Low level of qualification	Migration outflow of the population
	Need for 89.9 thou. new workers by 2035	Population growth rate: 7%	Population growth rate: 3%	Need for 52.1 thou. new workers by 2035	Decrease in enrollment in Arctic universities and their branches: 71%	Population growth rate: 0.2%
T	Depletion of an easily accessible mineral resource base	Low balance of the transportation network	Depreciation of energy infrastructure	Underdeveloped energy connections	Depreciation of energy infrastructure	Complicated construction of transportation infrastructure
	Exploration cost in the structure of production projects: 30%	Growth rate of road density: 28%	Degree of depreciation of fixed assets: 47.1%	Total length of main power grids: 18,740.8 km	Degree of depreciation of fixed assets: 45.2%	Share of highways meeting regulatory requirements: 50.42%
E	Increase in waste volumes	Impact of objects of accumulated harm	Growing technogenic and anthropogenic impact by the mining sector	Secondary contamination of drinking water in water supply networks	Accidents at production sites	Permafrost thawing
	Growth rate of production and consumption waste generation: 30%	Number of objects (water areas) of accumulated harm: 13	Reduction in agricultural land areas by 27.8 thou. hectares	Degree of depreciation of fixed assets of water supply and wastewater disposal and treatment: 31.6%	Number of man-made emergencies in 2020: 2	Reduction in pile bearing capacity: 18%
L	Lack of specific technical legislation	Tightening of environmental quality standards	Lack of specific technical legislation	Establishment of new expenditure commitments	Tightening of environmental quality standards	Lack of consideration of the specifics of remote regions
	Number of spheres of legal regulation in the field of technical norming and standardization: 8	Physical volume index of environmental expenditures: 130.7	Number of spheres of legal regulation in the field of technical norming and standardization: 8	Income tax for companies producing and exporting LNG: 34%	Physical volume index of environmental expenditures: 160.2%	Timeframe for the implementation of energy infrastructure projects: 14 months

in Appendix C).

On the basis of Formulas (1)–(5), the ranking of companies by their presence in the AZRF was made (presented in Figure 5).

Among the considered domestic and foreign companies, the leaders in terms of the number of relevant cases were (in descending order) PJSC Gazprom, PJSC ROSSETI, SC Rosatom, Rosneft Oil Company, and LTD Emperium. A total of 218 cases from 15 technological blocks were analyzed, reflecting the stages of business operations and ensuring proper living conditions for the talent pool. Based on the results of the assessment, the two most significant cases were identified for each of the blocks in the ATPCs (90 cases out of the initial 218).

In total, 61 measures were formulated for six programs as a response to risk events. Of these, 25 measures responded directly to zonally distributed risks. Each of the Programs, in turn, was responsible for a certain sphere: “National Economy”, “Business”, “Science and Education”, “National Issues”, “Social Policy”, and “Environment”.

The authors do not provide a complete list of measures, as they were compiled for a particular state, taking into account all its peculiarities, and are not the key ones in this paper. When using the model proposed by the authors for forecasting the development of another state or conducting a repeated forecast for the Russian Federation, the list of measures should be formed separately.

The risk values X_n‴ were obtained within each ATPC for the three scenarios with their own value (108 risk values X_n‴, based on the initial 36 values of X_n). At the same time, for each risk, three to five groups of similar measures aimed at neutralizing it were selected.

As a result of the above, it is possible to obtain both the final values of cluster risks (presented in

Table A5. Quantitative values of principal risks in 2035 for Central ATPC.

Risk/ Scenario	Negative		Baseline		Positive
P	Transportation and logistical isolation	Technological isolation	Transportation and logistical isolation	Technological isolation	Transportation and logistical isolation	Technological isolation
	Number of railroad tracks from the “big land” to the region: 38	The level of availability of industry technology: 80%	Number of railroad tracks from the “big land” to the region: 41	The level of availability of industry technology: 135%	Number of railroad tracks from the “big land” to the region: 41	The level of availability of industry technology: 152.42%
E	Falling investment attractiveness and activity	Increase in expenditures for the repair and maintenance of facilities	Falling investment attractiveness and activity	Increase in expenditures for the repair and maintenance of facilities	Falling investment attractiveness and activity	Increase in expenditures for the repair and maintenance of facilities
	Growth rate of investment activity: 13.65%	Share of emergency housing stock: 3.67%	Growth rate of investment activity: 13.54%	Share of emergency housing stock: 0.99%	Growth rate of investment activity: 11.7%	Share of emergency housing stock: 0.61%
S	Migration outflow of the population	Low level of qualification	Migration outflow of the population	Low level of qualification	Migration outflow of the population	Low level of qualification
	Population growth rate: 8%	Need for 57.66 thou. new workers by 2035	Population growth rate: 12.67%	Need for 5.8 thou. new workers by 2035	Population growth rate: 15%	A surplus of 0.61 thou. new jobs through 2035
T	Depreciation of energy infrastructure	Underdeveloped energy connections	Depreciation of energy infrastructure	Underdeveloped energy connections	Depreciation of energy infrastructure	Underdeveloped energy connections
	Degree of depreciation of fixed assets: 47.1%	Total length of main power grids: 35,264.2 km	Degree of depreciation of fixed assets: 0%	Total length of main power grids: 41,310.9 km	Degree of depreciation of fixed assets: 0%	Total length of main power grids: 46,341.7 km
E	Growing technogenic and anthropogenic impact by the mining sector	Secondary contamination of drinking water in water supply networks	Growing technogenic and anthropogenic impact by the mining sector	Secondary contamination of drinking water in water supply networks	Growing technogenic and anthropogenic impact by the mining sector	Secondary contamination of drinking water in water supply networks
	Reduction in agricultural land areas by 27.8 thou. hectares	Degree of depreciation of fixed assets of water supply and wastewater disposal and treatment 7.6%	Reduction in agricultural land areas by 3.02 thou. hectares	Degree of depreciation of fixed assets of water supply and wastewater disposal and treatment 0%	Reduction in agricultural land areas by 3.02 thou. hectares	Degree of depreciation of fixed assets of water supply and wastewater disposal and treatment 0%
L	Lack of specific technical legislation	Establishment of new expenditure commitments	Lack of specific technical legislation	Establishment of new expenditure commitments	Lack of specific technical legislation	Establishment of new expenditure commitments
	Number of spheres of legal regulation in the field of technical norming and standardization: 8	Income tax for companies producing and exporting LNG: 11.93%	Number of spheres of legal regulation in the field of technical norming and standardization: 25	Income tax for companies producing and exporting LNG: 6.81%	Number of spheres of legal regulation in the field of technical norming and standardization: 25	Income tax for companies producing and exporting LNG: 11.93%

in Appendix D, with an example for the central ATPC) and the trend of changes in scenario values for the complex of clusters as a whole. Three scenarios can be analyzed at once, which results in an analytical forecast and formulation of response measures.

In the example under consideration, the negative scenario suggests an increase in the gap between developed and subsidized subjects of the AZRF. The total production increases up to 1389.18 t c.f./year, the trend of which (as well as other forecast values) is presented in Figure 6.

The total volume of NSR transportation by 2035 does not reach the target values (121.99 mln tons per year instead of 220 mln tons per year). Emphasis on the development of technologies required by foreign importers reduces the ratio of the number and capacity of icebreakers (to 0.0123, while the target is 0.0131 pcs./MW). The navigation period along the NSR (183 days/year) is not year-round and depends solely on climatic conditions. River and marine navigation continue to increase emissions (up to 73.9 mln tons of CO₂-eq./year) due to the absence of the LNG transition program [43].

The installed capacity of power plants in the AZRF due to the low implementation of large-scale infrastructure modernization programs is 8654.15 MW. The specific fuel consumption for electricity and the heat supply is 113.27 g c.f./kWh and 0.47 g c.f./Gcal, respectively.

Under the baseline scenario, the implementation of most strategies and programs proceeds normally. The opening of new markets (to APR) increases the total volume of transportation by 2035 (see Figure 7) compared to the negative scenario (up to 137.56 mln tons per year).

The increase in cargo turnover is stable, but not extensive, provided by the development of transportation (the growth of the ratio of the number and capacity of icebreakers reaches 0.0116 pcs./MW) and navigational technologies (the navigation period eventually amounts to 191 days/year). In river and sea transport, LNG-powered ships are gradually introduced (emission reduction to 74.14 mln tons of CO₂-eq./year). The total production to meet the needs of domestic consumers and importers amounts to 1420.17 t c.f./year.

The installed capacity of power plants in the AZRF is 8709.68 MW, with specific fuel consumption for electric power of 112.21 g c.f./kWh. The same indicator for heat energy is 0.45 g c.f./Gcal.

According to the positive scenario, the country’s trade potential is developed by competition between APR countries (India, China). The total production is increased up to 1432.35 t c.f./year. A significant share of these resources is used for domestic consumption and production.

In addition to export activities, transit through the NSR is partially implemented. The total volume by 2035 reaches 143 mln tons/year (see Figure 8).

The amount of the fleet (0.01 pcs./MW) develops in tandem with the principles of environmental friendliness (with the reduction in emissions to 74.22 mln tons of CO₂-eq./year due to LNG ships). The abovementioned has only a positive impact on the duration of navigation along the NSR: 194 days/year.

The installed capacity of power plants in the AZRF is 8735.2 MW. The specific fuel consumption for electricity and the heat supply is 111.73 g c.f./kWh and 0.45 g c.f./Gcal, respectively.

Thus, the scenario modeling of the impact of risk changes on KPIs (at the level of technology clusters) and the assessment of the direct impact of external factors on risks will make it possible to trace their mutual influence at the country level. The approach can use and vary both different technological solutions and programs of state-level measures that respond to risk phenomena. The risks themselves are determined by the specifics of the region under consideration.

5. Conclusions

In the framework of this study, the authors described the approach for a comprehensive assessment of the current economic, geopolitical, technological, and social potential of the AZRF. The proposed approach made it possible to model the development in the context of three scenarios from 2015 to 2035 by means of a structural combination of a wide range of data analysis methods. The model is based on statistical data, and standard proven methods of regression analysis are used to forecast them in dynamics. However, the results are modified in forecasting by taking into account expert opinions. They determine the degree of influence of external factors on KPIs and the relevance of risks based on their awareness of the current regional situation. At the same time, the risks of territorial development realized in the present do not depend on the authors. Since the selection of a correct expert group can be considered the basis for the reliability of the obtained results, the verification of the obtained values is difficult. However, the authors initially pursued the goal of obtaining indicative forecast target values, which should be aimed at in the opinion of the expert community. For this purpose, it is required, among other things, to implement specific infrastructure, economic, and social projects considered in the model that contribute to the achievement of KPIs.

The main forecast values obtained from the modeling have the following ranges in view of different scenarios:

-

The total volume of NSR transportation by 2035—121.99–143 mln tons per year (with a target of 220 mln tons per year);

-

The CO₂-equivalent emission by 2035—73.9–74.22 mln tons of CO₂-eq./year;

-

The navigation period along the NSR—183–194 days/year;

-

The total production—1389.18–1432.35 t c.f./year;

-

The specific fuel consumption for electric power—111.73–113.27 g c.f./kWh;

-

The specific fuel consumption for heat power—0.45–0.47 g c.f./Gcal;

-

The average ratio of the number and capacity of icebreakers—0.01–0.0123 pcs./MW (the target is 0.0131 pcs./MW).

The obtained ranges of changes in the forecast values characterize the vector in which the development of technologies and the area under consideration will take place under different conditions. The modeling results allow for assessing the sufficiency of measures taken to achieve the target indicators.

The approach proposed by the authors is universal, as it is suitable for analyzing the development of any complex and heterogeneous territory: the division into ATPCs and the determination of the main forecast values are expert in nature and easily adaptable. Building a model based on risk changes, in turn, reflects not only the changes that occur but also their causes. The authors emphasize that the proposed and described model is not a final product predicting the state of the Arctic Zone by 2035 but one of the possible tools for the formation of and change in public policy in matters of innovative development.