A Dynamic Analysis to Examine Regional Development in the Context of a Digitally Enabled Regional Innovation System: The Case of Western and Central Macedonia (Greece)

Abstract

The significance of Regional Innovation Systems (RIS) as a strategic tool for enhancing a region’s competitiveness has been increasingly recognized. This paper presents a model of RIS that was developed using the system dynamics (SD) methodology. The goal of this model is to amalgamate the systemic approach with computer modeling and simulation disciplines into a comprehensive dynamic framework for analyzing RIS. Within this framework, the paper explores the impact of smart technologies on regional development through the RIS. Specifically, the SD model serves as an ‘experimental tool’ for conducting extensive what-if scenario analyses concerning smart technologies. The efficacy of these technologies is examined in terms of their dynamic influence on regional development, with insights derived from simulation outcomes. Data from two Greek regions provides a strategic analysis over a designated time horizon.

1. Introduction

1.1. Definitions and Context

Innovation is a core driver of national social development. It plays a key role in enhancing the competitive advantages of organizations [1]. With global technological advancements and rapid industrial evolution, scientific and technological innovation has become the foundation of national economic competitiveness. Due to the development of economic globalization, the competition for innovation between different regions has evolved into competition between regional systems [2]. Consequently, a series of regional innovation centers have emerged in different countries. In this context, improving regional innovation capabilities and designing scientific and efficient innovative development models play crucial roles in the current economic development of developed as well as developing countries.

Regional innovation refers to the process of generating new ideas, technologies, and practices within a specific geographic area, typically with the aim of fostering economic growth, competitiveness, and social development. It encompasses various factors of the regional ecosystem, such as research and development (R&D), entrepreneurship, collaboration between academia and industry, infrastructure, and supportive policies tailored to the particular regional characteristics.

In addition to collaboration, infrastructure plays a crucial role in enabling regional innovation. Access to physical infrastructure like transportation networks, communication technologies, and research facilities is essential for facilitating the exchange of ideas and knowledge. Shenzhen, China, provides an example of a region that has transformed from a manufacturing center into a leading innovation hub driven by its focus on technology, entrepreneurship, and government support [3].

However, regional innovation also faces challenges. Access to funding for R&D and startup ventures can be limited in regions with fewer financial resources. Talent retention is another concern, as regions must compete to retain skilled workers and prevent a “brain drain” to more established innovation hubs. Moreover, effective policy coordination among government agencies, industry stakeholders, and educational institutions is essential to create an enabling environment for innovation [4].

Overall, regional innovation is a multifaceted process involving collaboration, infrastructure, entrepreneurship, and supportive policies. By leveraging their unique strengths and addressing challenges, regions can enhance their competitiveness, spur economic growth, and improve quality of life [5,6,7].

1.2. Components of the Regional Innovation System

From the standpoint of innovation systems, a Regional Innovation System (RIS) is an open system made up of innovation resources (technology, knowledge, and information) [8] and innovation actors (government, core businesses, universities, research institutes, and intermediaries) [9,10]. The innovation actors in a regional innovation system foster the diffusion, dissemination, and innovation of knowledge within the region [10,11] under the combined influence of innovation environments (i.e., market environment and social environment), thereby enhancing the evolution and efficiency of knowledge diffusion in the system.

Regional innovation systems are no longer limited by time or location in a digital context. Instead, they reorganize the current innovation resources and processes to create links and interactions across innovation actors, relying on digital technologies, knowledge-sharing platforms, crucial complementary resources, and knowledge-driven innovation.

Overall, the approach put forward in the literature and adopted by this paper views the RIS as comprising six subsystems, which include (i) a Subsystem of Capacity in Information and Communication Technologies (ICT), (ii) a Subsystem of Innovation and Regional Development; (iii) a Subsystem of Institutional Framework; (iv) a Subsystem of Knowledge Implementation/Capitalization; (v) a Subsystem of Knowledge Networking; and (vi) a Subsystem of Knowledge Production/Dissemination. These are briefly explained below.

The characteristics of a Regional Innovation System (RIS) pertaining to contemporary Information Technology (IT), its application, and people’s IT-related abilities are referred to as the Subsystem of Competence in ICT. By bringing individuals together who share interests, encouraging collaborative knowledge generation, and fostering the growth of intelligent learning environments, ICT facilitates the spread of information [12].

The Innovation and Regional Development subsystem includes processes such as product innovation, process innovation, strategic intelligence, cooperative R&D, the emergence of new markets, the drawing of knowledge-intensive industries, and the production of spin-offs [13].

The subsystem of the Institutional Framework focuses on the role of institutions in innovation. Inside an institutional framework, a variety of private and governmental organizations interact and function together. The traditions, customs, human value systems, and social structure of a nation are frequently referred to as the “institutional environment” or “institutional framework” [13]. Particular laws and norms, as well as national and international regulations, make up the institutional framework at the regional level [14].

In the subsystem of Knowledge Implementation/Capitalization, knowledge application and exploitation constitute a significant portion of RISs’ business operations. This subsystem’s primary constituents include clients, contractors, partners, and competitors [15]. A dynamic regional innovation ecosystem revolves around creative enterprises and clusters that possess strong learning capacities and can convert their existing knowledge into commercial success [16].

The Subsystem of Knowledge Networking refers to how supply chain relationships, firms, and networks of practices can create and/or spread knowledge through interaction [17]. Knowledge can be categorized into explicit and tacit forms. “Explicit knowledge” refers to information that is documented, conveyed using formal languages, and stored in databases, archives, and libraries. Conversely, “tacit” information is difficult to formalize and convey through other human communication channels due to its individualized character. The accumulation of innovative activity is facilitated by tacit knowledge that is based on location [5].

Finally, the subsystem of Knowledge Production/Dissemination is concerned with the sharing and transferring of knowledge through formal and informal practices. Knowledge transfer can be acquired through an established knowledge network, which is viewed as a network of connections between actors that facilitates learning between businesses and organizations [18]. An economy’s knowledge base can be described as its capacity to generate and invent novel concepts, ideas, procedures, and goods, as well as how they might be interpreted in relation to economic growth [19].

1.3. Literature Review and Research Aims

Regions play a crucial role in the global economy, being the first to experience the impact of economic changes and thus actively engage in the formulation of research and innovation policies [13]. The Regional Innovation System (RIS) is a key tool for implementing these policies and analyzing the innovation process at the regional level. Since the early 1990s, the European Commission has introduced various policy programs that provide a strategic view of technology and innovation at the regional level [20]. Programs such as the Regional Infrastructures and Strategies for Innovation and Technology Transfer (RITTS), Regional Technology Plans (RTPs), and Regional Innovation Strategies (RIS) have offered co-funding and guidance to regional governments. These programs aim to assess regional innovation potential and define strategies that promote cooperation and capacity building among small businesses, the research and technology community, and public authorities, ultimately achieving regional development [21].

The issue of regional development in Greece has posed a consistent obstacle due to economic imbalances, unequal allocation of resources, and differing degrees of infrastructural development. To address these issues, European strategies such as Regional Innovation Strategies (RIS) and the Entrepreneurial Discovery Process (EDP) have played crucial roles [22]. The Entrepreneurial Discovery Process (EDP) is an iterative process that involves stakeholders from various sectors to identify and exploit new opportunities for innovation and growth. EDP is a cornerstone of the smart specialization strategy, which aims to prioritize investments in key areas where regions have competitive advantages [23]. These strategies are particularly evident in the regional policy frameworks and implementations, such as those seen in the region of Attica, Greece [24].

The development strategy (2014–2020) includes Research and Innovation Strategies for Smart Specialization (RIS3). Based on the principles of the Europe 2020 strategy for smart, sustainable, and inclusive growth, RIS3 aims to achieve high levels of employment, productivity, and social cohesion within the EU and its Member States. Smart specialization is vital for sustainable growth, offering opportunities in both domestic and global markets. It also contributes to inclusive growth among regions, strengthens territorial cohesion, and manages structural changes, creating jobs and fostering social innovation [22].

Additionally, the concept of Smart Cities has gained prominence. Globally, smart cities are seen as fundamental to the sustainable development of urban centers, ensuring their future expansion under sustainable conditions. Smart cities are part of the broader objective of Western societies, which is to transition towards a knowledge-based economy and society. These cities enhance human capacities for creativity, learning, and innovation. According to Komninos [25], smart cities comprise three key components:

1. The innovation system (local/regional) of the reference region guides the development of knowledge and technologies within regional organizations (businesses, universities, technology centers, incubators, etc.).

2. Smart Technologies, including digital information and knowledge management applications that facilitate communication, decision-making, technology transfer, and collaboration in innovation.

3. The reference area, or the city itself, consists of the physical space and its inhabitants.

This research explores how the components of smart cities are linked at the regional level. The literature review identifies specific gaps, particularly in the complexity of policy planning and smart specialization, which involve multiple stakeholders and often insufficient information. Ranga and Etzkowitz [26] highlighted the need for strengthened multi-level policies that require comprehensive evidence to identify and select regional priorities effectively [27].

To address these gaps in strategy development, skills, and methods, the literature suggests leveraging Smart Technologies. Digital platforms are proposed as a solution, providing a dynamic economic structure that can enhance stakeholder engagement and the use of advanced datasets. These platforms enable the creation of ecosystems where users can collaborate across a wide range of activities [28,29,30]. Such environments can be used for dissemination and sharing common goals, enhancing collaboration among stakeholders, and guiding innovation during the RIS3 design process [25,31].

Innovation, collaboration, and coordination can be effectively developed through network relationships [32]. Consequently, online platforms in policy formulation and strategic planning should be considered key components of the Regional Innovation System [25].

This research focuses on the use of New Technologies in Regional Innovation Systems for Regional Development. To this end, we will create a dynamic model capturing the level of Regional Development in relation to the contribution of smart technologies by the actors involved in the Regional Innovation System. This model will be applied to two Greek regions with different Regional Development indicators: the Region of Western Macedonia and the Region of Central Macedonia. It will serve as a guide and methodological tool for formulating appropriate regional policies by identifying strengths and weaknesses within the Regional System and enhancing those that provide multiplier benefits for Regional Development.

The research aims to answer whether the use of new technologies in a Regional Innovation System benefits Regional Development, as defined and measurable by specific indicators. The study focuses on modeling the influence of intelligent technologies on regional development, employing the RIS framework and utilizing the system dynamics technique. The objective is to examine the dynamic impacts of these technologies and offer a strategic analysis based on scenarios within two distinct Greek regions. Specifically, the study analyzed different possible values for:

• The percentage of home Internet connections;
• The presence of ICT specialists in the study region;
• The usage of personal computers; and
• The extent of ICT programs in the study area.

The model indirectly differentiates the values of the remaining independent variables.

1.4. Overview of the Study Regions

The mathematical model we present allows the empirical evaluation of the operation of a regional innovation system. The model developed is applied to the RISs of two Greek NUTS 2-level regions, Western and Central Macedonia, and various scenarios of changes in the indicators of smart technologies are developed to observe their impact on the two regions under consideration’s regional development.

The Region of Western Macedonia is located in the northwest of the country. It is the only region of Greece that is located far from the sea and is not on main tourist routes. It covers an area of 9451 km² and has a relatively small population of 250,4531 (2023), which is decreasing due to immigration. The capital of the region is Kozani. The other much smaller population centers are Ptolemaida, Grevena, Florina and Kastoria. The region’s GDP per capita amounts to EUR 5.549262 million PPS (2022), representing only 60% of the EU average [33].

Western Macedonia is a threatened region due to its high dependence on fossil fuel-related industry. For several decades, Western Macedonia was the main energy-producing region in Greece. Home to around 80% of Greece’s lignite industry, the region provided well over 70% of the country’s grid electricity at its peak—resulting in high CO₂ emissions and negative health impacts from associated air and water pollution. In the face of increasingly strict EU environmental directives, lignite power generation in the region has been phased out since 2010 [34].

Economic decline and unemployment are the main socioeconomic challenges for the region. Unemployment rates in Western Macedonia hovered around 30% in 2012–2018 and have fallen to around 23% in recent years (16.7% in 20233) as people leave the region to find work elsewhere (Eurostat). Youth unemployment in the region is among the highest in the EU. It is estimated that over 20% of regional businesses have ceased operations since 2008, while the turnover of those that remain, particularly in the trade sector, has fallen by 40% [35].

The region has a weak innovation potential, which is exacerbated by a strong outflow of people, especially among young people. There is almost no private R&D investment. The region’s other main sectors—fur industry and agri-food—are not traditionally knowledge-based. Overall, Western Macedonia—an emerging innovative actor according to the EU’s regional innovation scoreboard [36]—currently does not present a fertile environment for innovation-driven growth [37].

Western Macedonia is leveraging smart technologies to enhance public services, environmental management and energy efficiency. Urban areas implement smart city projects to improve mobility, waste management, and energy efficiency, which support the goals of sustainable and equitable urban development [33,38].

In addition, the region is exploring the application of digital technology in agriculture (precision farming), tourism (digital platforms to improve tourist experiences) and natural resource management. The aim is to utilize ICT to promote economic development, improve the quality of life and maintain environmental sustainability.

Central Macedonia, situated in northern Greece, encompasses the major urban center of Thessaloniki, as well as smaller cities such as Katerini and Serres. Its economy is characterized by diversity, with key sectors including industry, agriculture, tourism, and services [39].

Central Macedonia is a modest innovator. Innovation performance has increased over time [36]. Central Macedonia is a vibrant and diverse region located in the northern part of Greece. It is a focal point in terms of geography, economy, and culture. It is the second most populous region in Greece, and it includes the city of Thessaloniki, which is the second largest city in the country. Thessaloniki itself is a center of cultural activities, historical sites and educational institutions, making it a focal point for tourists and residents alike.

From an economic point of view, Central Macedonia is a power in the Greek context. It has a dynamic economy with diverse sectors such as agriculture, industry and services. The region is an important agricultural producer known for its fruits, vegetables and wine, thanks to its fertile plains. Industrial activity is also important, with an emphasis on areas such as food processing, textiles and heavy industry in the Thessaloniki area [39].

The region boasts a robust industrial base, with manufacturing activities spanning textiles, food processing, chemicals, pharmaceuticals, and automotive sectors. Thessaloniki stands out as a major industrial and commercial hub, attracting businesses and fostering economic activity [40].

Agriculture thrives in Central Macedonia due to its fertile land and favorable climate. The region produces grains, fruits, vegetables, olives, and tobacco, contributing significantly to its economy. Tourism is another vital sector, drawing visitors with its cultural heritage, historical sites, and scenic landscapes. Thessaloniki offers Byzantine monuments and a vibrant cultural scene, while coastal resorts along the Thermaic Gulf and Halkidiki peninsula attract beachgoers and tourists. Services play a crucial role, with Thessaloniki serving as a commercial and financial center, hosting businesses, banks, and service-oriented enterprises. The city’s port and transportation infrastructure facilitate trade and connectivity within the region and beyond.

Central Macedonia faces challenges like unemployment and the need for innovation. However, opportunities for growth exist, including further development of tourism, promotion of entrepreneurship, and leveraging its strategic location. In summary, Central Macedonia’s economy is diverse, supported by its industrial strength, agricultural resources, tourism appeal, and strategic location, positioning it as a significant contributor to Greece’s economic landscape [40].

2. System and Problem Description

2.1. System Description

Figure 1 exhibits a very generic description of the RIS under study. In particular, the conceptualization of the regional innovation system as comprising six subsystems is adopted in light of the aforementioned literature review, focusing on regional innovation systems and the innovative processes occurring within them, prior to evaluating and examining the differences among the Greek regional systems. The subsystems themselves consist of various parts, each of which can be quantitatively assessed. Due to the dynamic nature of the system and its subsystems, the characteristics of each factor impact the others [24]. The subsystems include:

• The ICT capacity subsystem.
• The innovation and regional development subsystem considers innovative outcomes.
• The institutional framework subsystem includes regional governance.
• The knowledge capitalization and application subsystem (comprising businesses and clusters).
• The knowledge network subsystem.
• The knowledge production and dissemination subsystem (involving universities and research centers).

Figure 1. The RIS structure.

Figure 1. The RIS structure.

As illustrated in Figure 1, the value of the constitutive variable RIS, which represents the RIS, is influenced by the RIS growth rate (R_in RIS) and the temporal depreciation of the RIS (R_out RIS). The R_in RIS rate is derived as the average of the change rates of the subsystems composing the RIS. Specifically, it is affected by the following variables:

• R_in ICT: Growth rate of the ICT capacity subsystem;
• R_in TID: Growth rate of the innovation and regional development subsystem;
• R_in IE: Growth rate of the institutional framework subsystem;
• R_in AEoK: Growth rate of the knowledge capitalization and application subsystem;
• R_in NoK: Growth rate of the knowledge network subsystem;
• R_in PDoK: Growth rate of the knowledge production and dissemination subsystem.

Conversely, the decrease rate R_out RIS refers to the depreciation affecting the entirety of the content of the subsystems that constitute the RIS over the years.

The analysis of the individual subsystems with the help of influence diagrams has been published by Samara et al. [41]. In this paper, we present the mathematical formulation of the model and consider alternative scenarios in terms of the % of home Internet connections, the existence of ICT specialists in the study area, the use of PCs, and the level of existence of ICT programs in the study area. The values of the remaining independent variables are indirectly varied through the model.

2.2. Problem Description

The mathematical model presented allows for the empirical evaluation of the performance of a regional innovation system. The developed model is applied to the RIS of two Greek regions, Western and Central Macedonia, and various scenarios involving changes in the indicators of smart technologies are developed to observe their impact on the regional development of the two regions under examination.

3. The SD Model

Stocks and flows, together with feedback loops, constitute the foundational elements of System Dynamics (SD) theory. In this framework, stocks represent accumulations captured as the net result of inflows and outflows within a system. SD employs a specific diagrammatic notation for representing these elements, where stocks are depicted as rectangles, inflows are denoted by arrows entering stocks, and outflows by arrows exiting stocks [42]. Stocks are quantified through stock equations, which are essentially the time integrals of net flows.

Flows, on the other hand, are characterized by rate equations, which describe them as time-dependent functions of stocks and other system parameters. Within SD models, the portrayal of stocks and flows allows for the representation of time as a continuous variable, facilitating the occurrence of events and changes at any continuous point in time. These stock-flow dynamics are typically converted into a system of differential equations for computational analysis, with simulations generally conducted using sophisticated graphical simulation software such as PowerSim 10^®, Vensim^®, i-think^®, and Stella^® [43,44].

The comprehensive stock-flow diagram of the SD model presented here includes 20 stock variables, 39 rate variables, and 131 auxiliary variables. This section aims to elucidate the mathematical modeling involved in such analyses, striving to maintain a level of generality in the description.

3.1. Generic Stock and Flow Structure

The generic stock and flow diagram for the two subsystems—Innovation and Regional Development and ICT Capacity—are displayed in Figure 2a–c, due to the size of the model; the remaining stock and flow diagrams are provided in Appendix A.

The scoring scale for the Innovation and Regional Development subsystem is as follows:

• 1 → subsystem level incomplete;
• 7 → excellent.

The performance level of the subsystem is determined by the constitutive variable named Territorial Innovation and Development (TID), which is increased by the R_in TID growth rate and decreased by the R_out TID decay rate, the latter being due to the depreciation of subsystem achievements over time.

The R_in TID rate is increased by the TID inflow variable, divided by the time T TID, which describes the interval required for all factors to alter the subsystem’s profile. The TID inflow variable is calculated as the minimum value between the difference between the actual state of the innovation and regional development subsystem and its desired value (with 7 being the maximum) and the average rating of the factors influencing it, including:

• Product Innovation (PI), which increases according to the number of researchers, the percentage of those contributing to R&D, and the proportion involved in PI (see Figure 2b).
• Process Innovation (PI), which changes according to the PI growth rate, influenced by a delay function and the number of researchers engaged in PI (similar to Product Innovation). Both PI and Process Innovation decrease according to a depreciation rate (Depreciation of PI and Process Innovation, respectively), attributed to the knowledge degradation over time (see Figure 2b).
• Public sector R&D expenditures at the NUTS 2 regional level (GOV) (GERD).
• The employment rate of the age group 20–64 by NUTS 2 regions (TID1) describes the regional employment rate of the age group 20–64 as a percentage of the population in the same age group.
• The NUTS 2 unemployment rate representing the unemployed as a percentage of the economically active population (i.e., the labor force or the sum of the employed and unemployed).

The average value of the factors affecting this subsystem (TID average) is obtained by assigning the value of each of them on a scale of 1 to 7. This process is carried out by means of variables called “Grade <variable name>”.

In a manner analogous to the Innovation and Regional Development subsystem scale, the subsystem’s capacity in ICT is graded on a scale from 1 to 7, where 1 indicates an inadequate level and 7 represents an optimal level. The performance of the subsystem is determined by the constitutional variable ICT, which increases due to the R_in ICT rate and decreases due to the R_out ICT rate, the latter being attributed to the depreciation over time.

The R_in ICT rate is augmented by the ICT inflow variable, which is divided by the time T TID. This period describes the interval required for all influencing factors to modify the subsystem’s profile. The ICT inflow variable is calculated as the minimum value between the difference between the actual state of the subsystem and its desired state (7 being the maximum) and the average rating of the factors affecting the subsystem under analysis. These influencing factors include:

• Businesses employing ICT specialists (National A1I), a factor directly dependent on the GERD index.
• Individuals with more than basic general digital skills (National C2), where the number decreases with an increase in centrally dictated processes and increases when these are reduced.
• Digitization—Longitudinal Evolution (National A1W), a factor affected by the per capita value of the GERD index.
• Percentage of businesses with e-commerce sales (National A1A), which correlates with the population aged 25–34 who have completed tertiary education.
• Cloud computing services (National A1F) related to the number of researchers engaged in R&D activities.

The average value of the factors influencing this subsystem (ICT average) is derived by mapping each factor’s value to the 1–7 scale through variables named ‘Grade <variable name>’.

3.2. Equations

Innovation and Regional Development subsystem

(a) Stock equations:

$$\mathrm{process} \mathrm{innovation}\left(t\right)=\mathrm{process} \mathrm{innovation}\left(\mathrm{t}=0\right)+{\int }_0^t\left[Rin PROCIN\left(t\right)-Depreciation of PROCIN\left(t\right)\right] \mathrm{dt} [\%]$$

(1)

$$\mathrm{product} \mathrm{innovation}\left(t\right)=\mathrm{product} \mathrm{innovation}\left(\mathrm{t}=0\right)+{\int }_0^t\left[Rin PI\left(t\right)-Depreciation of PI\left(t\right)\right] \mathrm{dt} [\%]$$

(2)

$$\mathrm{Total} \mathrm{Innovation} \mathrm{and} \text{Development-TID}\left(t\right)=\mathrm{Total} \mathrm{Innovation} \mathrm{and} \text{Development-TID}\left(\mathrm{t}=0\right)+{\int }_0^t\left[Rin TID\left(t\right)-Rout TID\left(t\right)\right] \mathrm{dt} \left[\mathrm{undisturbed}\right]$$

(3)

(b) Flow equations:

$$Depreciation of PI(t)=DELAYMTR(R_{in}\mathrm{P}\mathrm{I};2;3;0) \left[\frac{\%}{year}\right]$$

(4)

$$Depreciation of PROCIN(t)=DELAYMTR(R_{in}\mathrm{P}\mathrm{R}\mathrm{O}\mathrm{C}\mathrm{I}\mathrm{N};2;3;0) \left[\frac{\%}{year}\right]$$

(5)

$$\begin{array}{c}{R}_{in}\mathrm{P}\mathrm{I}\left(t\right)=MIN(Discrepancy of PI(t)/T_{pi};Product innovation per researcher\ast RD capacity(t)\\ \ast percentage of researchers for PI\ast \% ICT(t)) \left[{\displaystyle \frac{\%}{year}}\right]\end{array}$$

(6)

$$\begin{array}{c}{R}_{in}\mathrm{P}\mathrm{R}\mathrm{O}\mathrm{C}\mathrm{I}\mathrm{N}\left(t\right)=MIN(Discrepancy of PROCIN(t)\\ /T_{pri};MIN(PI to PROCIN\left(t\right);\left(1-percentage of researchers for PI\right)\\ \ast Process innovation per researcher\ast RD capacity\left(t\right)\ast \%ICT) \left[{\displaystyle \frac{\%}{year}}\right]\end{array}$$

(7)

$$R_{in}\mathrm{T}\mathrm{I}\mathrm{D}\left(t\right)=DELAYPPL(TID inflow;T{ }_{TID};0)$$

(8)

$$R_{out}\mathrm{T}\mathrm{I}\mathrm{D}\left(t\right)=DELAYMTR(R_{in}\mathrm{T}\mathrm{I}\mathrm{D};2;3;0) \left[\frac{1}{year}\right]$$

(9)

(c) Auxiliary equations:

$$\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{c}\mathrm{r}\mathrm{e}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y} \mathrm{o}\mathrm{f} \mathrm{P}\mathrm{I}\left(\mathrm{t}\right)=\mathrm{M}\mathrm{A}\mathrm{X}(0;\mathrm{m}\mathrm{a}\mathrm{x} \mathrm{P}\mathrm{I}-\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t} \mathrm{i}\mathrm{n}\mathrm{n}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}(t\left)\right) [\%]$$

(10)

$$\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{c}\mathrm{r}\mathrm{e}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y} \mathrm{o}\mathrm{f} \mathrm{P}\mathrm{R}\mathrm{O}\mathrm{C}\mathrm{I}\mathrm{N}\left(\mathrm{t}\right)=\mathrm{M}\mathrm{A}\mathrm{X}(0;\mathrm{M}\mathrm{A}\mathrm{X}(0;\mathrm{M}\mathrm{A}\mathrm{X} \mathrm{P}\mathrm{R}\mathrm{O}\mathrm{C}\mathrm{I}\mathrm{N}-\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s} \mathrm{i}\mathrm{n}\mathrm{n}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(t\right)\left)\right) [\%]$$

(11)

$$\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{c}\mathrm{r}\mathrm{e}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y} \mathrm{o}\mathrm{f} \mathrm{T}\mathrm{I}\mathrm{D}\left(\mathrm{t}\right)=\mathrm{O}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{a}\mathrm{l} \mathrm{T}\mathrm{I}\mathrm{D}-\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{I}\mathrm{n}\mathrm{n}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{a}\mathrm{n}\mathrm{d} \mathrm{D}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t} \left(\mathrm{T}\mathrm{I}\mathrm{D}\right)\left(t\right) \left[\mathrm{undisturbed}\right]$$

(12)

$$Grade of GERD\left(\mathrm{t}\right)\left(\mathrm{C}entral Macedonia\right)=\mathrm{G}\mathrm{R}\mathrm{A}\mathrm{P}\mathrm{H}\left(\mathrm{G}\mathrm{E}\mathrm{R}\mathrm{D}\right(t);7;5;\{1;2;3;4;5;6;7\left\}\right)\left[\mathrm{undisturbed}\right]$$

(13)

$$Grade of GERD\left(\mathrm{t}\right)(\mathrm{W}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{n} \mathrm{M}\mathrm{a}\mathrm{c}\mathrm{e}\mathrm{d}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{a})=\mathrm{G}\mathrm{R}\mathrm{A}\mathrm{P}\mathrm{H}\left(\mathrm{G}\mathrm{E}\mathrm{R}\mathrm{D}\right(t);\mathrm{5,02};\mathrm{1,5};\{1;2;3;4;5;6;7\left\}\right)\left[ \mathrm{undisturbed}\right]$$

(14)

$$\begin{array}{c}Grade of NUTS 2 unemployment rate\left(\mathrm{t}\right)\\ =\mathrm{G}\mathrm{R}\mathrm{A}\mathrm{P}\mathrm{H}\left(\mathrm{N}\mathrm{U}\mathrm{T}\mathrm{S} 2 \mathrm{u}\mathrm{n}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\left(t\right);16;\mathrm{2,5};\left\{7;6;5;4;3;2;1\right\}\right)\left[\mathrm{undisturbed}\right]\end{array}$$

(15)

ICT subsystem

(a) Stock equations:

$$\mathrm{ICT}(t)=\mathrm{ICT}(\mathrm{t}=0)+{\int }_0^t\left[R_{in}\mathrm{I}\mathrm{C}\mathrm{T}\left(t\right)-R_{out}\mathrm{I}\mathrm{C}\mathrm{T}\left(t\right)\right] \mathrm{dt} [\mathrm{undisturbed}]$$

(16)

(b) Flow equations:

$$R_{in}\mathrm{I}\mathrm{C}\mathrm{T}\left(t\right)=DELAYPPL\left(ICTinflow;Tict;0\right)$$

(17)

$$R_{out}\mathrm{I}\mathrm{C}\mathrm{T}\left(t\right)=DELAYMTR(R_{in}\mathrm{I}\mathrm{C}\mathrm{T};2;3;0) \left[\frac{1}{year}\right]$$

(18)

(c) Auxiliary equations:

$$\begin{array}{c}\mathrm{\%} \mathrm{o}\mathrm{f} \mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{u}\mathrm{n}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{t}\mathrm{y} \mathrm{s}\mathrm{t}\mathrm{u}\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{s}\left(\mathrm{t}\right)\\ =(\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{u}\mathrm{n}\mathrm{i} \mathrm{s}\mathrm{t}\mathrm{u}\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{s} \ast \mathrm{\%} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} 25-34\_\mathrm{n}\mathrm{o}\mathrm{n} \mathrm{r}\mathrm{e}\mathrm{t}\mathrm{i}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t})/(\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\\ \ast \mathrm{\%} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} 25-34)\left[\mathrm{\%}\right]\end{array}$$

(19)

$$\mathrm{B}\mathrm{u}\mathrm{s}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{e}\mathrm{s} \mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{I}\mathrm{C}\mathrm{T} \mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{s}-\mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{I}\left(\mathrm{t}\right)=\mathrm{G}\mathrm{E}\mathrm{R}\mathrm{D}\left(t\right)\ast \mathrm{K}\mathrm{A}1\mathrm{I}2\ast \mathrm{K}\mathrm{A}1\mathrm{I}1\left[\mathrm{\%}\right]$$

(20)

$$\mathrm{D}\mathrm{i}\mathrm{g}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{z}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}-\mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{W}\left(\mathrm{t}\right)=\mathrm{K}\mathrm{A}1\mathrm{W}1\ast \mathrm{G}\mathrm{E}\mathrm{R}\mathrm{D} \mathrm{p}\mathrm{e}\mathrm{r} \mathrm{h}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{t}\mathrm{a}\mathrm{t}\left(t\right)\ast 1000-\mathrm{K}\mathrm{A}1\mathrm{W}2\left[\mathrm{\%}\right]$$

(21)

$$\mathrm{D}\mathrm{i}\mathrm{s}\mathrm{c}\mathrm{r}\mathrm{e}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y} \mathrm{o}\mathrm{f} \mathrm{I}\mathrm{C}\mathrm{T}\left(\mathrm{t}\right)=\mathrm{M}\mathrm{A}\mathrm{X}(0;\mathrm{o}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{a}\mathrm{l} \mathrm{I}\mathrm{C}\mathrm{T}-\mathrm{I}\mathrm{C}\mathrm{T}(\mathrm{t})\left[\mathrm{undisturbed}\right]$$

(22)

$$\mathrm{I}\mathrm{C}\mathrm{T}1 \mathrm{a}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e}\left(\mathrm{t}\right)=\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\left(\begin{array}{c}\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{A}\left(t\right);\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{F}\left(t\right);\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{I}\left(t\right);\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{W}\left(t\right);\\ \mathrm{G}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}nal A1W\left(t\right);Grade of C2\left(t\right) \left[\mathrm{u}\mathrm{n}\mathrm{d}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{e}\mathrm{d}\right]\end{array}\right)$$

(23)

$$\mathrm{I}\mathrm{C}\mathrm{T}2 \mathrm{a}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e}\left(\mathrm{t}\right)=\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\left(ICT specialists;PCusage;internet usage\left(t\right);ICT projects\right)\left[\mathrm{undisturbed}\right]$$

(24)

$$\begin{array}{c}\mathrm{I}\mathrm{C}\mathrm{T} \mathrm{i}\mathrm{n}\mathrm{f}\mathrm{l}\mathrm{o}\mathrm{w}\left(t\right)=MIN({\displaystyle \frac{AVERAGE\left(ICT1 average \left(t\right);ICT2 average \left(t\right)\right)}{Tict}};{\displaystyle \frac{discrepancy of ICT\left(t\right)}{Tict}}\\ +target adaptation ICT(t)\left[\mathrm{undisturbed}\right]\end{array}$$

(25)

$$\begin{array}{l}\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{N}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{A}1\mathrm{A}\left(t\right)=IF(National A1A(t)<\mathrm{0,08};1;IF(National A1A\left(t\right)\\ <\mathrm{0,14};2;IF(National A1A(t)<\mathrm{0,16};3;IF(National A1A\left(t\right)\\ <\mathrm{0,18};4;IF(National A1A(t)<\mathrm{0,20};5;IF(National A1A\left(t\right)\\ <\mathrm{0,22};6;7\left)\right)\left)\right)\left)\right)\left[\mathrm{undisturbed}\right]\end{array}$$

(26)

$$\begin{array}{l}Grade National A1F\left(t\right)=IF(National A1F(t)<\mathrm{0,10};1;IF(National A1F\left(t\right)\\ <\mathrm{0,12};2;IF(National A1F(t)<\mathrm{0,14};3;IF(National A1F\left(t\right)\\ <\mathrm{0,16};4;IF(National A1F(t)<\mathrm{0,18};5;IF(National A1F\left(t\right)\\ <\mathrm{0,22};6;7\left)\right)\left)\right)\left)\right) \left[\mathrm{undisturbed}\right]\end{array}$$

(27)

$$\begin{array}{l}Grade National A1I\left(t\right)=IF(Businesses employing ICT specialists-National A1I(t)\\ <\mathrm{1,79};1;IF(Businesses employing ICT specialists-National A1I(t)\\ <\mathrm{1,87};2;IF(Businesses employing ICT specialists-National A1I(t)\\ <\mathrm{1,96};3;IF(Businesses employing ICT specialists-National A1I(t)\\ <\mathrm{2,04};4;IF(Businesses employing ICT specialists-National A1I(t)\\ <\mathrm{2,13};5;IF(Businesses employing ICT specialists-National A1I(t)\\ <\mathrm{2,30};6;7\left)\right)\left)\right)\left)\right) \left[\mathrm{undisturbed}\right] \left(\Sigma \chi έ\sigma \eta \right)\end{array}$$

(28)

$$\begin{array}{l}Grade National A1W \left(t\right)=IF(Digitization-National A1W(t)\\ <\mathrm{48,09};1;IF(Digitization-National A1W(t)\\ <\mathrm{54,1};2;IF(Digitization-National A1W(t)\\ <\mathrm{60,11};3;IF(Digitization-National A1W(t)\\ <\mathrm{66,13};4;IF(Digitization-National A1W(t)\\ <\mathrm{72,14};5;IF(Digitization-National A1W(t)\\ <\mathrm{84,16};6;7\left)\right)\left)\right)\left)\right) \left[\mathrm{undisturbed}\right]\end{array}$$

(29)

$$\begin{array}{l}Grade of C2 \left(t\right)=IF(People with above basic general digital skills-C2(t)\\ <\mathrm{15,5};IF(People with above basic general digital skills-C2(t)\\ <\mathrm{16,8};2;IF(People with above basic general digital skills(t)-C2\\ <18;3;IF(People with above basic general digital skills (t)-C2\\ <\mathrm{19,3};4;IF(People with above basic general digital skills (t)-C2\\ <20;5;IF(People with above basic general digital skills \left(t\right)-C2\\ <23;6;7\left)\right)\left)\right)\left)\right) \left[\mathrm{undisturbed}\right]\end{array}$$

(30)

$$\begin{array}{l}internet usage \left(t\right)=IF(\%home internet connection\\ <\mathrm{0,14};1;IF(\%home internet connection\\ <\mathrm{0,28};2;IF(\%home internet connection\\ <\mathrm{0,42};3;IF(\%home internet connection\\ <\mathrm{0,56};4;IF(\%home internet connection\\ <\mathrm{0,70};5;IF\left(\%home internet connection<\mathrm{0,84};6;7\right)\left)\right)\left)\right))\end{array}$$

(31)

$$National A1A\left(t\right)=KA1A\ast \% of graduated university students \left(t\right)\left[\%\right]$$

(32)

$$National A1F\left(t\right)=RD capacity\left(t\right)\ast KA1F1+KA1F2 \left[\%\right]$$

(33)

$$xPeople with above basic general digital skills-C2\left(t\right)=-4082\ast number of procedures+\mathrm{35,246} [thousand people]$$

(34)

$$target adaptation ICT\left(t\right)=DELAYINF(Rout ICT(t);2;1;0)$$

(35)

4. Numerical Experimentation and Discussion

In order to study the impact of Smart Technologies on Regional Development, alternative price scenarios were examined in terms of the percentage of household Internet connections, the existence of ICT specialists in the area under study, the use of PCs, as well as the level of existence of ICT programs in the area under study. The values of the remaining independent variables are indirectly varied through the model.

Table 1. Alternative Parameter Values.

	Worst-Case					Best-Case Scenario
% household internet connections	20	40		60		80	100
ICT specialists in the area	1	2	3	4	5	6	7
PC usage	1	2	3	4	5	6	7
Existence of ICT programs	1	2	3	4	5	6	7

shows the different values each of these variables takes, starting from the worst to the best case.

Therefore, the total number of simulations carried out is equal to 7 × 7 × 7 × 5 = 1715 simulations for each region, i.e., 3430 simulations for both regions. Each simulation had a period of 10 years (120 months) with a time step of 0.2 years. The simulation starts in 2024, and findings from the two simulation groups, one for each region under consideration, are presented upon completion of the simulation process.

The purpose of the specific simulations was to identify, for each region separately, the combinations under study of factors of smart technologies to achieve faster development of Regional Development. For this reason, those combinations were sought for which the value of calculating the level of the Regional Development subsystem (TID) balances at the highest values and in the shortest possible time.

Also, the combinations of the parameters under study for which the influence of the subsystem of smart technologies on the subsystem of Regional Development is stronger are sought. For this purpose, the TID_ICT_impact indicator was created, whose value is equal to:

$${\mathrm{T}\mathrm{I}\mathrm{D}\_\mathrm{I}\mathrm{C}\mathrm{T}}_{impact}=\frac{TID}{ICT}$$

The higher the value of the specific index, the smaller the impact of smart technologies on Regional Development, and the opposite if the value of the index takes smaller values. Also, when the said indicator receives values smaller than unity, then for each unit of development of the smart technologies’ subsystem (on a scale of 1–7), the development of the Regional Development subsystem receives values greater than unity, while the reverse occurs when the TID_ICT_impact index takes values greater than unity.

Listed below are the worst and best possible combinations of alternative values of the independent variables under investigation, as depicted in Table 1, for each control area separately. In this way, the parameters that should be given special attention are identified, as well as those that are of secondary importance.

4.1. Region of Central Macedonia

To identify the parameters that are critical for the highest possible Regional Development and the maximum possible impact of the smart technologies’ subsystem on it, 1715 simulations were performed with all possible combinations of their values. Table 1 shows the best-case and worst-case scenario combinations of the studied parameters of smart technologies in terms of Regional Development.

The TID_ICT_impact index is used to identify the best and worst combinations. In particular, the case for which its value is as small as possible and, if possible, less than unity, as well as the fluctuation of the index values, is considered optimal. On the other hand, the cases in which there is a large fluctuation in the index values, and its values are even higher than the unit, are considered the worst. For this reason,

Table 2. Best-case scenario combinations (Central Macedonia).

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Best-Case Combination (B.C.)	20	5	2	7	0.9908	0.2682	1.11
		4	3	7
	40	5	6	2	0.9908	0.2682	1.11
	60	4	6	1	0.9908	0.2682	1.11
	80	3	4	3	0.9908	0.2682	1.11
		2	5	3
	100	4	3	2	0.9908	0.2682	1.11

and

Table 3. Worst-case scenario combinations (Central Macedonia).

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Worst-Case Combination (W.C.)	20	1	1	1	1.7646	0.4897	2.04
	40				1.6409	0.45911	1.91
	60				1.4390	0.4036	1.67
	80				1.3560	0.3811	1.57
	100				1.2817	0.3595	1.49

show the best and worst combinations, respectively, for each possible value of the % of home internet connections, both the variation in the index values and the average value, as well as the maximum value. In addition, Figure 3 and Figure 4 show indicative values of the index for the best and worst combination in the case that the percentage of home internet connections equals 20% and 60%, respectively.

The results demonstrate that, as expected, the worst combinations that result refer to cases where all the factors are in zero growth, as they receive the worst possible value. What is worth noting is the large variation in the maximum value of the TID_ICT_impact index as the percentage of home internet connections decreases from 60% to 20%, compared to the case where there is a decrease from 100% to 60%. We also observe that even in the worst combinations, if the level of the percentage of home internet connections is high, then the average value of the impact index (TID_ICT_impact) of smart technologies in Regional Development is relatively satisfactory as it is close to unity and the variation in its values is satisfactory.

As far as the optimal combinations are concerned, we notice that in all cases, the mode of operation of the system is the same as both the value of the maximum value and the average value, as well as the variation in the values of the influence index, are the same. This fact is also reflected in Figure 5, where the values obtained by the Territorial Development (TID) calculation variable for the optimal combination for each of the possible values of the percentage of home internet connections are collectively noted. We observe that for all cases, the optimal combination has the same impact on the value of Regional Development.

Furthermore, from Table 2, it is observed that in order to be able to achieve the optimal combination if the percentage of home internet connections is low (20% or 40%), there should be the development of ICT programs to the maximum extent possible (7), while the existence of ICT specialists in the area and the use of PCs can vary at medium levels. On the other hand, when the percentage of connections increases, then the existence of ICT programs becomes a factor of secondary importance, and the use of PCs and the existence of ICT specialists in the area now play a primary role. In particular, the use of a PC is considered more important in cases where the percentage of home connections ranges between 40%, 60%, and 80%, while the existence of ICT specialists in the area is more critical in cases where the % of home internet connections reach 100%. It is worth noting that the effect of smart technologies on Regional Development is quite high, and in cases where all factors receive the maximum values. However, the best maximum value of the index is not achieved during the simulation horizon, i.e., the case in which it receives the smallest possible value but also the lowest variation.

4.2. Region of Western Macedonia

Following a similar methodology as in the case of the Region of Central Macedonia, an attempt is also made for the Region of Western Macedonia to identify the parameters that are critical in order to achieve the best possible Regional Development, with the maximum possible impact of the smart technologies’ subsystem on it. For this purpose, 1715 simulations were carried out for all possible combinations of the values of the smart technologies’ parameters under consideration, as shown in Table 1.

After the completion of all the simulations, the results were examined in order to identify the best and worst possible combinations of the studied parameters of smart technologies in terms of Regional Development.

To identify the best and worst combinations, use is made, as in the previous case, of the TID_ICT_impact indicator. In particular, the case for which its value is as low as possible and, if possible, less than unity is considered optimal, as is the variation in the index values. On the other hand, cases in which there is a large fluctuation of the index values, and its values are high, are considered worse, and even worse when they are higher than unity.

Table 4. Best-case scenario combinations (Western Macedonia) for the percentage of household internet connections = 20%.

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Best-Case Combination (B.C.)	20	1	7	7	1.0004	0.1859	1.08
		2	7	6
		4	5	6
			4	7
		5	7	3
			5	5
		6	7	2
			3	6
		7	7	1
			5	3
			3	5
			2	6
			1	7

,

Table 5. Best-case scenario combinations (Western Macedonia) for the percentage of household internet connections = 40%.

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Best-Case Combination (B.C.)	40	1	7	6	1.0004	0.1859	1.08
		2	5	7
		3	7	4
			6	5
			5	6
		4	5	5
		5	7	2
			4	5
			3	6
			2	7
		6	5	3
			3	5
		7	4	3
			1	6

,

Table 6. Best-case scenario combinations (Western Macedonia) for the percentage of household internet connections = 60%.

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Best-Case Combination (B.C.)	60	2	6	4	1.0004	0.1859	1.08
		3	7	2
			6	3
			4	5
			3	6
			2	7
		4	4	4
		5	5	2
			3	4

,

Table 7. Best-case scenario combinations (Western Macedonia) for the percentage of household internet connections = 80%, 100%.

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Best-Case Combination (B.C.)	80	1	4	6	1.0004	0.1859	1.08
			3	7
		2	5	4
		3	4	4
			2	6
			1	7
		4	2	5
			1	6
		5	3	3
			2	4
		6	4	1
			1	4
		7	1	3
	100	7	1	2	1.0004	0.1859	1.08

and

Table 8. Worst-case scenario combinations (Western Macedonia).

	% Household Internet Connections	ICT Specialists in Area	PC Usage	Existence of ICT Programs	TID_ICTimpact
					Average Value	Variance	Max. Value
Worst-Case Combination (W.C.)	20	1	1	1	1.8333	0.3058	2.02
	40				1.7027	0.2853	1.88
	60				1.4904	0.2520	1.65
	80				1.4030	0.2384	1.55
	100				1.2817	0.3595	1.49

show, respectively, the best and worst combinations for each possible value of the % of home internet connections, considering both the variation in the index values and the average value as well as its maximum value. Also, in Figure 6 and Figure 7, the values of the index for the best and worst combination are shown indicatively in the case where the percentage of home internet connections is equal to 40% and 80%, respectively.

The results demonstrate that, as expected, the worst combinations that result refer to cases where all the other factors are in zero growth, as they receive the worst possible value. What is worth noting is the large variation in the maximum value of the TID_ICT_impact index as the percentage of home internet connections decreases from 60% to 20%, compared to the case where there is a decrease from 100% to 60%. Also, it is perceived that in the case of Western Macedonia, the variation in the index values is smaller than in the case of Central Macedonia. Also, for the impact index (TID_ICT_impact) to obtain a relatively satisfactory value in the case of the worst combinations, the percentage of home internet connections should reach the maximum. Otherwise, the influence of smart technologies on Regional Development is incomplete, even in the case that the percentage of household internet connections reaches 80%.

As far as the optimal combinations are concerned, we notice that in all cases, the mode of operation of the system is the same as both the value of the maximum value and the average value, as well as the variation in the values of the influence index, are the same. This fact is also reflected in Figure 8, where the values obtained by the Territorial Innovation and Development (TID) calculation variable for the optimal combinations for each of the possible values of the percentage of home internet connections are collectively noted. We observe that in all cases, the optimal combination has the same impact on the value of Regional Development.

Also, in the case of Western Macedonia, it is noticed that there is more variety in the optimal combinations when the percentage of household internet connections is small, while this variety decreases as this percentage increases. This fact indicates the flexibility that exists in policymaking in order to increase the impact of smart technologies on Regional Development.

It is worth noting that the existence of ICT programs is an important factor as they should range at a medium or high level, as seen in the majority of optimal combinations. Also, as the results state, both the average value and the maximum value of the optimal combinations are slightly greater than unity, which means that smart technologies will have an impact on Regional Development equal to their system change. On the other hand, we see that the variation in the influence index is noticeably lower compared to the case of Central Macedonia.

4.3. Policy Implications

Based on the findings from the simulations, we can make the following recommendations for policymakers in the regions of Western and Central Macedonia:

• Western Macedonia

Given the region’s historical dependence on lignite and the consequent economic decline, diversification is crucial. The simulation results indicate that ICT infrastructure is vital for regional development. A broadband expansion by accelerating the rollout of high-speed internet across the region, particularly in rural and underserved areas, to ensure comprehensive digital connectivity is considered a key policy for Western Macedonia. By increasing the share of residential and business internet connections, access to digital services and opportunities for education and entrepreneurship can be improved. This could potentially contribute to the development of innovation and the local economy. Moreover, implementing smart city projects in urban areas like Kozani and Ptolemaida to improve public services, transportation, and energy management through the use of IoT and AI technologies will be of high importance to smart city initiatives. Considering the region is undergoing an energy transition, and there will be a large number of the population that will need upskilling, launching extensive training programs to upskill the workforce in ICT and digital skills is a priority for the region. To do so, collaboration with local universities and vocational schools should take place to provide courses tailored to industry needs.

• Central Macedonia

Based on the simulation results for the case of Central Macedonia, the implementation of the targeted policies below can significantly enhance regional development. More precisely, Central Macedonia can capitalize on the region’s diverse industrial sectors and boost innovation by promoting advanced manufacturing techniques and Industry 4.0 technologies in key sectors like textiles, food processing, and chemicals and by fostering collaboration between universities, research institutes, and industries to drive innovation and technological advancement.

Furthermore, Central Macedonia could leverage ICT to boost regional development. This can be accomplished by implementing widespread digital literacy and ICT training programs targeting both young people and the existing workforce. Taking into consideration that agriculture thrives in Central Macedonia, enhancing precision agriculture could offer multiplier benefits for regional development. Finally, supporting innovative startups and technology-based businesses is of critical importance for Central Macedonia. This could be achieved by offering incentives and aids for the creation and development of new ICT startups that would strengthen the local economy and create new jobs.

By adopting these proposals, both regions can achieve stronger, more sustainable and dynamic growth by exploiting the potential of smart technologies.

5. Conclusions, Limitations and Future Research

This study uses a System Dynamics model of a Regional Innovation System (RIS) to clarify the important role that smart technologies play in promoting regional growth. The results highlight how smart technology can impact regional innovation and competitiveness in a dynamic way. The study offers actual evidence to support the claim that incorporating smart technologies into regional innovation systems can stimulate innovation and economic growth by applying the model to two Greek regions. The developed scenarios show how areas can effectively change their economic landscapes by combining innovation policy with the correct technical infrastructure.

The results show that, in the case of Central Macedonia, the percentage of home internet connections is low, and the development of ICT programs to the maximum extent possible is crucial. On the other hand, if this percentage is high, then the use of PCs and the existence of ICT specialists in the area play a primary role. In the case of Western Macedonia, it is observed that while the percentage of household internet connections is low, there is greater variation in the best combinations; however, as this percentage increases, the variety decreases. In addition, it is noticeable that the majority of best combinations indicate that ICT programs should be at a medium or high level; therefore, their existence is a significant consideration.

Although this study offers insightful information about how smart technologies affect regional development, it must be noted that it has several limitations. First off, the study’s scope is limited to just two Greek regions, which means it might not be representative of other geographic contexts with various socioeconomic features. Furthermore, the reliability of the results may be impacted by the accuracy and accessibility of the regional data, especially in domains with constantly advancing technology. A survey of the literature found a startling dearth of regional data. The authors “regionalized” national-level indicators using a novel analytical approach to make up for this [45]. Using correlations and regressions in a “model fit” approach, they estimated regional-level values for four indicators that are not available there, based on the values of four other indicators that are available there and are intimately related to the former group. These were cloud computing, digitalization, employed ICT professionals, and e-commerce sales.

To generalize the results, future research should concentrate on broadening the study’s geographical scope to encompass a variety of regions with different technological and economic environments. To improve the model’s relevance and prediction ability, real-time data and more dynamic policy variables should be added. A deeper understanding of the strategic planning of regional innovation systems may be gained by examining the long-term effects of smart technologies on district socioeconomic variables and high-effect indicators, including advanced business models [46,47], innovative entrepreneurship types [48], smart education [49], and public welfare within the framework of the Regional Innovation Systems.