Technological Innovation and the Role of Smart Surveys in the Industrial Context

Abstract

Technological innovation has significantly transformed the field of statistics, not only in data analysis but also in data collection. Traditional methods based on direct observation have evolved into hybrid approaches that combine passively collected data (e.g., from GPS or accelerometers) with active user input through digital interfaces. This evolution has led to Smart Surveys—next-generation tools that leverage smart devices, such as smartphones and wearables, to collect data actively (via questionnaires or images) and passively (via embedded sensors). Smart Surveys offer strategic value in industrial contexts by enabling real-time data collection on worker behavior, environments, and operational conditions. However, the heterogeneity of such data poses challenges in management, integration, and quality assurance. This study proposes a modular system architecture incorporating gamification elements to enhance user participation, particularly among hard-to-reach worker segments, such as mobile or shift workers. By leveraging motivational strategies and interactive feedback mechanisms, the system seeks to foster greater engagement while addressing critical data security and privacy concerns within industrial Internet of Things (IoT) environments.

1. Introduction

The digital evolution has had a pervasive impact across all sectors of society, with the convergence of advanced technologies such as the Internet of Things (IoT) and Artificial Intelligence (AI) radically transforming the way individuals and organizations interact with their surrounding environment [1]. IoT devices, equipped with sensors capable of interpreting context and communicating with each other through connected networks [2], enable data collection in multiple application areas, which, if properly leveraged, can support decision-making processes. The global number of IoT devices is estimated to increase from 19.8 billion in 2025 to over 40.6 billion by 2034, with approximately 60% represented by the consumer segment [3]. The potential of IoT has significantly expanded thanks to its convergence with machine learning and Artificial Intelligence, which allow the analysis of large volumes of sensor data to understand real-world problems better and make critical operational decisions. To address complex challenges and meet computing and communication requirements, IoT and machine learning must integrate and mutually support each other, thus increasing the value of data generated by distributed devices. IoT technologies are increasingly employed in logistics and transportation management in the industrial sector, aiming to improve customer experience and reduce operational costs [4]. Complementarily, the industrial Internet of Things plays a fundamental role in the development of real-time smart factories [5], inventory management [6], and customer satisfaction [7]. Despite the widespread adoption of IoT in various sectors, the literature has yet to document a direct and systematic use of these technologies to implement Smart Survey (SS). However, the latter presents broad potential applications [8]. Digital devices used in Smart Surveys can collect heterogeneous data via internal sensors (GPS modules, cameras, motion sensors integrated into mobile devices) and external sensors (interaction with smartwatches, smart climate control systems, etc.). The data obtained from these functionalities, called smart data, can be processed in real time during collection, directly involving the respondent, and afterwards. Processing operations, defined as smart tasks, can be aggregated into innovative services or solutions accessible via the survey application interface. The development and greater accessibility of IoT devices can revolutionize data collection systems, including surveys, allowing real-time monitoring of human behavioral and environmental parameters and offering new opportunities to obtain more precise and contextually relevant data [9]. In this context, sensors play a central role, enabling IoT devices to acquire detailed information through various functionalities. These include the detection of electromagnetic signals (such as Bluetooth, GPS, GSM, NFC, Wi-Fi), audio recording [10], motion measurement through accelerometers and gyroscopes [11], and the collection of environmental data such as temperature, humidity, atmospheric pressure, and magnetic fields [12]. Additionally, biometric and proximity sensors [13] allow for the collection of personal data, such as fingerprints and heart rate, while visual sensors capture images and videos for advanced analyses. Data collected in real time and interconnected through the IoT network allow Smart Surveys to overcome the limitations of traditional data collection methodologies. Integrating internal and external sensors in devices used by respondents enriches self-reported information with objective and contextual data. Moreover, the local processing capabilities of smart devices facilitate the implementation of smart tasks, improving the quality and relevance of collected data. Consequently, IoT expands data collection capabilities and transforms Smart Surveys into a dynamic, efficient, and highly innovative tool for contemporary statistical research. This is particularly true for studies analyzing data in industrial contexts [14], such as assessing employee well-being in workplaces. The main barriers to their widespread adoption are ethical and regulatory, since passive data collection is only possible with informed consent [15]. To overcome these limitations, this proposal introduces an innovative survey system that uses gamification as a motivational lever to increase voluntary and conscious participation in Smart Surveys in industrial contexts. Through game mechanics (badges, levels, symbolic rewards), the goal is to transform the data collection experience into an engaging activity and promote continuous participation. The originality of this proposal focuses on two key aspects:

• This literature review explores the topic of SSs to identify its key characteristics and potential, focusing on possible applications in the industrial sector.
• A system enabling SSs in business settings is presented. The proposed solution allows data collection through smart devices (e.g., wearables, smartphones) and environmental sensors, using methods that do not interfere with regular operational activities.
• The strategic use of gamification as a motivational tool aims to engage employees actively, overcoming the negative or passive perception often associated with traditional internal surveys. The goal is to enable and promote the widespread adoption of SSs, ensuring continuous and sustained participation over time.

The remainder of this paper is organized as follows. Section 2 introduces the foundational concepts of SSs and provides a comprehensive review of the relevant literature. Section 3 examines the key challenges of implementing SSs in industrial contexts. Section 4 presents the architecture of a developed scenario that illustrates how SSs, supported by IoT-enabled data collection, can enhance data utilization within industrial environments. Section 5 illustrates, through a simulation, a possible application of the proposed solution. The concluding Section 8 summarizes the main findings and outlines potential limitations and directions for future research.

2. The State of the Art on Smart Surveys

Most surveys rely on traditional methodologies for statistical research, such as interviews, observations, focus groups, online surveys, experiments, and paper-based questionnaires. Currently, the most widely used data collection method is the survey. This approach involves sending a standardized questionnaire, which includes closed or open questions, to a selected sample of respondents. Surveys are commonly employed in descriptive, explanatory, and exploratory research. Moreover, they are particularly effective for reaching a large number of participants and for posing questions consistently and systematically [16].

Although the above-mentioned techniques contribute significantly to research, they are often characterized by several limitations. For example, online surveys, typically distributed via Internet platforms, do not allow researchers to gather background information about the population to which the survey is presented. Moreover, these surveys may be subject to self-selection bias, in which only specific individuals choose to respond [17]. Other traditional methods, such as interviews and experiments, face high costs, long durations, and difficulties accessing and analyzing the collected data, mainly when conducted offline.

In contrast, SSs represent an advanced data collection form integrating interactive features and innovative technologies to contextualize responses. These surveys leverage the capabilities of modern digital devices to enhance both the quality and depth of the data collected. Smart features may include internal sensors (e.g., GPS, motion sensors, cameras), external devices (e.g., smartwatches, environmental sensors), and access to online public data sources such as OpenStreetMap [18]. Additionally, with appropriate consent, SSs can request access to personal data—such as bank or loyalty card records—directly from respondents or existing datasets held by the survey organization. Smart devices can also process information in real time by utilizing their built-in computing power, for example, by applying machine learning algorithms to recognize images or detect motion patterns. The data derived from these smart features, known as smart data, often requires further processing to become usable. In the case of raw sensor inputs, this processing may involve interpretation or transformation to produce meaningful information. These operations, called smart tasks, can be performed during the survey or afterward and may be integrated into smart services or solutions available through the survey interface. Overall, SSs represent a shift toward richer, more contextualized, and technologically enhanced data collection methods. They actively engage respondents while leveraging the advanced capabilities of modern devices.

To better illustrate the advantages of SSs, a comparative table has been developed (

Table 1. Comparison between different data collection methods and survey approach.

Methods	Real Time	Accessibility	Coverage	Engagement	Response Rate	IoT
Traditional Survey				✓	✓
Focus Group	✓			✓	✓
Observation	✓
Interview	✓				✓
Email Survey			✓
Online Survey		✓	✓	✓	✓
Smart Survey	✓	✓	✓	✓	✓	✓

), which contrasts them with traditional data collection methods. Specifically, the comparison includes the main tools commonly used in data collection: traditional surveys, focus groups, observations, interviews, email surveys, and online surveys. A series of fundamental survey requirements was selected for comparison, as also highlighted by Istat [19]. To be effective, a survey must ensure accessibility, considering the various characteristics of users, such as vulnerable groups (e.g., visually impaired individuals) [16]. In addition, it must guarantee a good level of coverage, that is, it should reach a representative sample of the target population [20]. Another essential requirement is the ability to collect data in real time, especially in specific application contexts where real-time information can significantly influence results [21]. Finally, the participation and response rate of the respondents are key parameters for practical evaluation and design of a survey. Engagement is crucial to ensuring high-quality responses, as motivated and interested respondents tend to provide more accurate and complete data [22,23]. Similarly, the response rate—the percentage of participants who complete the survey—is essential to guaranteeing sample representativeness and to reducing bias due to non-response [24].

The table presents a comparative analysis of various data collection methodologies. It is evident that only a subset of these methodologies—namely, Focus Group, Observation, Interview, and Smart Survey—enable real-time data collection, thereby conferring a notable advantage in immediacy and responsiveness to the phenomena under investigation. In contrast, Traditional Surveys and Email Surveys lack this capability, likely due to their asynchronous nature and administration modalities.

Regarding accessibility, online surveys and smart surveys emerge as markedly more accessible approaches, attributable to their ability to engage a broad audience without geographical or temporal limitations. Conversely, traditional methodologies and Focus Groups exhibit reduced accessibility, potentially due to logistical constraints and limitations in participant availability.

Regarding coverage, understood as the capacity to reach a representative and extensive sample, digital-based methods (Online Survey and Smart Survey) demonstrate superior efficacy. By comparison, traditional methodologies and Focus Groups display more restricted reach, which may be ascribed to operational and sampling limitations.

Engagement and response rate constitute critical determinants of data quality. The data indicate that Smart Survey, Online Survey, Focus Group, and Observation methodologies perform favorably concerning these factors, suggesting that interactivity and contextualization contribute to enhanced participant involvement. In contrast, Email Surveys are characterized by comparatively lower engagement, plausibly due to their less interactive and more standardized format.

Finally, concerning integrating Internet of Things (IoT) devices, only Smart Surveys incorporate such technologies, highlighting their role as an advanced methodological innovation. This integration facilitates more precise, contextualized, and dynamic data collection using connected sensors and devices. In summary, the Smart Survey represents the most comprehensive and innovative data collection methodology, combining the benefits of real-time data acquisition, high accessibility, broad coverage, and elevated engagement and response rates, alongside seamless integration with IoT technologies. These attributes collectively enable Smart Surveys to surmount numerous limitations inherent to traditional digital survey methods.

Methodology

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [25] framework plays a critical role in developing review articles. It supports systematic reviews by promoting transparency in meta-analyses, ensuring transparent reporting of objectives, and drawing reliable and relevant conclusions from study findings. Consequently, this review adopts the PRISMA approach for conducting a comprehensive review of the literature on the SS, as depicted in Figure 1. The research materials used in this study were sourced from distinguished scientific journals covering the period from 2019 to March 2025, with searches conducted through the Google Scholar platform. A detailed compilation of scholarly articles was carefully chosen from renowned academic databases, including IEEE Xplore, Springer Link, ScienceDirect, arXiv, ACM Digital Library, ResearchGate, MDPI, and more. An initial selection of 4000 papers was made based on their relevance to the subject matter and specified keywords. A specific combination of search queries and keywords was utilized to conduct a thorough literature search for “Smart Survey”. The keywords used were as follows: “Smart Survey AND IoT OR sensor”; “Smart Survey AND IoT OR sensor”. The PRISMA protocols and guidelines were applied to select the final research papers to define a standardized set of inclusion criteria (IC) and exclusion criteria (EC). The inclusion criteria detail the requirements that the studies must satisfy to be included, while the exclusion criteria identify those that fall outside the scope of the review. This manual screening process improves the transparency and rigor of the selection procedure.

Table 2. List of inclusion criteria.

Inclusion Criteria	List of Inclusion Criteria
IC 1	Contains the term “Smart Survey” in the keywords, abstract, or titles.
IC 2	Published between 2019 and May 2025.
IC 3	Published in English.

and

Table 3. List of exclusion criteria.

Exclusion Criteria	List of Exclusion Criteria
EC 1	Results where the terms “smart” and “survey” appear separately and not as the compound expression “Smart Survey” in the title, abstract, or keywords.
EC 2	Results that refer to “Smart Survey” only as a software application or tool name.
EC 3	Results not relevant to the specific focus of the study.

summarize the inclusion and exclusion criteria established for this study.

The results of the PRISMA approach used to conduct a comprehensive literature review on Smart Surveys are presented in Figure 1.

Looking at the yearly trend of published articles on Smart Surveys (SSs), Figure 2 shows that the number of publications peaked in 2023. This increase reflects the growing interest in implementing SSs, driven partly by several European Union policies initiated since 2020.

The European Statistical System (ESS) launched pilot projects, called ESSNet, to explore the use of Smart Surveys (SSs), with particular focus on methodological, legal, and ethical aspects [26]. The ESSNet on Smart Survey (2020–2022) was a preparatory effort to develop a methodological and architectural framework for sharing and reusing innovative solutions and components, thereby supporting National Statistical Institutes (NSIs) in conducting SSs.

ESSNet on Smart Survey Implementation (SSI) 2023–2025: a new initiative aiming to operationalize the framework developed in the prior project, specifically within the social survey domain. This will primarily involve time use and household expenditure surveys, through field-tested, end-to-end process solutions that incorporate

• Active citizen participation as data contributors;
• Acquisition, processing, and combination of data gathered through smart device apps and sensor-acquired data;
• Ensuring the reliability of the final output by guaranteeing stringent personal data protection.

The initiative has also involved national statistical institutes, including ISTAT, which has recently contributed to the debate by highlighting how concerns related to privacy and data security may hinder participation, especially among older generations [18]. The promotion and launch of the project sparked interest in the academic world, with an increase in the number of published articles. In 2023, a peak was recorded, which may be linked to the start of the ESSNet on SS Implementation (SSI) 2023–2025 project.

The analysis of application contexts, as illustrated in Figure 3, highlights that despite the considerable potential of Smart Surveys (SSs), their use in industrial settings remains largely underexplored within the scientific literature. A systematic review of the relevant sources enabled the quantification of published contributions across different application domains. Specifically, the figure presents the distribution of studies across various disciplinary fields. The majority of research has focused on the healthcare sector, followed by the economic, technological, and information technology (IT) domains. Conversely, a smaller portion of the literature addresses political science, behavioral sciences, statistics, and the social sciences.

With regard to industrial applications, the growing interest in the adoption of SSs arises from their potential integration into the paradigms of Industry 4.0. In this context, it becomes feasible to leverage the capabilities of big data, intelligent machines, and the Internet of Things (IoT) to enhance both the effectiveness and efficiency of production processes, as well as to improve employee working conditions [14]. Accordingly, the application of SSs in industrial settings—similarly to other sectors—still offers significant potential for further research and innovation. An additional critical area of consideration concerns the protection of privacy and the security of personal data, both of which are becoming increasingly relevant with the continued expansion of artificial intelligence and big data technologies.

3. Smart Surveys in Industrial Contexts

In contemporary organizational settings, data collection practices rely predominantly on traditional methods, such as paper questionnaires or face-to-face interviews. These methods entail substantial costs and resource allocation and are often perceived by employees as marginal, low-impact activities, which may result in disengagement or minimal participation. This represents a considerable limitation, as gathering information about organizational well-being and job satisfaction is crucial for strategically managing human resources and improving the work environment. Knowing the mood of its employees allows a company to undertake targeted actions that can strengthen engagement and thus contribute to the overall improvement of the organization. Numerous empirical studies suggest that a high level of employee participation can positively affect both individual performance and organizational effectiveness. The most widely accepted definition in the literature is that of Schaufeli et al. [27], who describe engagement as “a positive, fulfilling, work-related state of mind characterized by vigor, dedication, and absorption. Rather than being a momentary and specific state, engagement is a more persistent and pervasive affective-cognitive condition not directed at any particular object, event, individual, or behavior.” However, a thorough review of the literature reveals that in more than 60% of studies, engagement is measured through subjective self-assessments [28], presenting several methodological limitations, such as social desirability bias, infrequent updates, and inconsistent participation. To address these challenges, SSs offer a promising approach. These hybrid monitoring systems combine active methods (e.g., voluntary input via digital interfaces) with passive data collection (e.g., automatic tracking via wearable devices, smartphones, and environmental sensors). IoT technologies, in particular, enable continuous, non-invasive, real-time measurement of variables related to employee well-being and engagement [29]. Therefore, integrating SSs with IoT technologies represents a significant opportunity to overcome the limitations of traditional data collection methods, improving the quality and timeliness of the information gathered. This innovative approach enables companies to more effectively monitor employee well-being and engagement, promoting more targeted and timely interventions that contribute to the continuous improvement of the work environment and overall growth of the organization. To facilitate the adoption of these solutions and overcome cultural resistance, the use of gamification techniques proves strategic, as it increases motivation, interest [30], and active participation of the users involved [31,32,33]. Applying gamified strategies to SSs not only improves response rates and the quality of the data collected but also strengthens employees’ sense of belonging, establishing the basis for a more inclusive, responsive, and innovation-oriented organization. SSs enable the transition from reactive to proactive data collection and, when integrated with gamification techniques, foster a profound cultural transformation where technology and behavioral sciences converge to support employee-centered innovation. This approach helps redefine the traditional relationship between employees and organizations, fostering a more participatory, transparent, and trust-based dynamic. Better aligning individual needs with organizational goals ultimately enhances employee well-being and overall business performance regarding productivity, efficiency, and effectiveness.

3.1. Gamification and Smart Surveys

Games have always been a fundamental component of human experience, serving as a form of social interaction and a tool capable of fostering engagement through immersive and rewarding experiences. From this premise, gamification is defined as “the use of game design elements in non-game contexts” [34]. Another definition, which highlights the relational and experiential dimension, describes it as “a process that offers affordances for playful experiences, aiming to facilitate value creation by the user” [35]. At the core of gamification is leveraging motivational drivers to increase user participation and performance across various contexts. Regardless of the application domain, gamification relies on typical game elements such as points, badges for achievements, challenges to overcome, and leaderboards, which encourage competition, interaction, and continuity in the proposed activity.

According to the review by Gupta and Gomtha [36], the concept of gamification began to spread around 2010, progressively extending to multiple fields. In particular, in corporate contexts, gamification is employed for various purposes: it boosts employee motivation and engagement and is used in training to enhance learning through techniques like points, badges, and challenges [37] and in personnel selection, where interactive business simulations stimulate candidate autonomy and motivation [38]. In knowledge management systems, it fosters collaboration and recognition through mechanics such as rewards, status, and leaderboards [39]. On corporate intranets and internal social networks, it stimulates participation and content sharing [40,41]. Finally, in crowdsourcing and ideation processes, it contributes to increasing creativity and involvement through competition dynamics and virtual rewards [42]. Beyond organizational settings, gamification has found significant applications in social research and data collection, particularly in online surveys. In the field of social sciences, various studies have shown how gamified systems can overcome one of the main limitations of traditional surveys, namely their perception as boring and impersonal activities. Some research has focused on the impact of different gamified survey design approaches [43,44]. For example, a study compared four types of questionnaires: a purely textual one, a textual one with decorative graphics, a functional one, and a fully gamified one. The results showed that the latter was the most effective regarding engagement, motivation, participation, and the quantity and quality of collected data. However, these results are not generalizable to all contexts. A systematic review by Hamari [35] emphasizes that the positive effects of gamification strongly depend on the context and characteristics of the users involved. This heterogeneity calls for further research to understand how and to what extent gamification influences the survey.

Hoekstra, in particular, distinguishes between psychological components, related to motivation and perceived enjoyment by respondents, and behavioral aspects, such as increased completion rates and improved response quality. He cites various studies confirming higher participation and completion in gamified surveys. However, he also mentions a study where a gamified textual version was completed less frequently than a traditional textual version [45]. This result has been criticized due to the lack of visual components and the inconsistency between the playful dynamic and the questionnaire content [46]. When visual elements and thematic connection with the survey topic are present, psychological and behavioral improvements are observed both psychologically and behaviorally [47]. Regarding the reliability and validity of data obtained from a gamified survey, Keusch et al. [46] highlighted how modifying a question to make it more “playful” could alter its meaning, compromising validity. Adding images or narratives could influence how the question is interpreted. Although the construction and validation of measurement instruments are well established, the issue of measurement error in gamified surveys remains underexplored. Keusch et al. [46] say that gamification may both increase and decrease measurement error, reflecting divergent opinions between critics and supporters: on one side, Dan and Lai [48] believe the playful environment may distract participants from the fundamental objective, creating a “playful and carefree” atmosphere; on the other, Turner et al. say that using gamification consistent with the survey theme can bring respondents closer to their daily experiences, encouraging more authentic and elaborate responses, especially in open-ended questions. In the context of SM, gamification can effectively increase user engagement in online surveys, thereby improving result accuracy. Besides making the completion experience more stimulating, it facilitates the adoption of technologies needed for data collection, such as installing IoT (Internet of Things) devices in monitored work environments. Playful mechanisms like rewards, challenges, or visual progress indicators could incentivize participants to actively complete the technical steps required, overcoming potential initial resistance. However, the literature analysis revealed a lack of studies explicitly linking gamification with SSs. No contributions directly exploring this integration were identified, highlighting the present work’s novelty and innovative potential.

3.2. Comparative Analysis and Practical Implications

Over time, numerous solutions integrating the Internet of Things and Gamification (IoTG) have been developed. The first to propose a combination of gamified elements and IoT technologies were Honig et al. [49], who, in 1985, introduced a rehabilitation system based on pressure sensors and television games. However, interest in IoTG applications has intensified mainly in recent decades, with implementations ranging from education to crowdsourcing and smart cities. The true strength of these solutions lies in their ability to enhance user engagement through gamified experiences that can stimulate intrinsic motivation. Furthermore, these systems are often designed to be intuitive and accessible even to users with heterogeneous technological skills [50]. Several studies have demonstrated the potential of gamification in promoting environmentally sustainable behaviors. In particular, many researchers have focused on collecting and analyzing data to predict users’ energy-related behaviors and optimize decision-making. For example, Konstantakopoulos et al. [51] proposed a gamified framework for smart infrastructures to encourage more responsible energy consumption. Franco [52], instead, integrated game mechanisms into multi-objective control strategies in ICT systems, balancing comfort and energy savings in public buildings. Beyond energy efficiency, more recent efforts have explored the use of IoTG solutions in workplace contexts to influence attitudes and behaviors. Iria et al. [53] tested a platform in an office building, but the lack of adequate sensors hindered automatic monitoring of energy loads, limiting the system’s effectiveness. This limitation was later addressed by Soares et al., who updated the building’s energy management system (BEMS), integrating it with sensors and real-time feedback mechanisms. Despite these advances, the literature on IoTG solutions remains relatively limited and predominantly focused on encouraging sustainable behaviors. Nevertheless, innovative proposals are emerging in other areas. For instance, Rocha et al. [54] presented a system for promoting physical activity through interactive games integrated with motion sensors. A recurring challenge in these systems is maintaining user interest and motivation over time. Recent studies [55,56,57] show that, despite initial enthusiasm, many users discontinue using gamified systems due to the repetitive nature of the activities. Moreover, the diversity of individual preferences renders one-size-fits-all approaches largely ineffective, highlighting the need for more personalized solutions. In this context, our proposal introduces important practical implications aimed at overcoming both methodological and technological limitations. On one hand, we apply the principles of IoTG to a still underexplored domain: Smart Surveys. On the other hand, we address the issue of repetitiveness by designing a gamification experience that is more varied, meaningful, and sustainable—even in complex, high-variability professional environments. Our gamification system aims to deliver a motivating experience focused on employee well-being and organizational efficiency. Integrating Smart Surveys and IoT technologies enables automated data collection, reducing the costs and complexity associated with traditional methods. Additionally, gamified components foster active participation and support more accurate and continuous monitoring of psychophysical states. Operationally, real-time monitoring allows for prompt responses to employee needs, fostering a supportive and collaborative work environment. This reduces stress levels, improves productivity, and lowers turnover rates. At the same time, adopting more conscious behaviors positively affects corporate energy consumption. Finally, the system’s ability to intelligently integrate biometric, environmental, and behavioral data enables early detection of critical situations—such as the onset of stress—and the deployment of targeted interventions. This approach ultimately improves the overall quality of work life and organizational performance.

4. System Architecture

The proposed system adopts a microservice-based architecture to enhance workplace environments by promoting employee well-being and optimizing energy efficiency (as shown in Figure 4). This is achieved by integrating intelligent sensing technologies, multimodal data acquisition, and AI-driven inference.

Biometric and behavioral data are continuously collected through wearable devices, such as smartwatches and smartphones, capturing heart rate variability, sleep quality, movement patterns, and stress indicators. In parallel, intelligent environmental sensors monitor contextual variables such as window state, lighting, Heating, Ventilation, and Air Conditioning (HVAC) usage, and standby status of the workstation, enabling the detection of energy-related behaviors and their correlation with well-being factors. All data streams converge into a neural network model implemented in Python (version 3.12.0), trained to extract multivariate temporal and contextual patterns. The model periodically infers user stress levels in real time by analyzing individual trends and deviations. Upon detecting elevated stress, the system triggers a dynamic intervention pipeline: personalized, context-aware SSs are generated to prompt reflective self-assessment. At the same time, gamified feedback mechanisms deliver real-time recommendations such as short breaks or light cognitive activities. These interventions are embedded in a gamification layer grounded in Self-Determination Theory, aimed at fostering intrinsic motivation through autonomy, competence, and relatedness. Users engage with micro-challenges and interactive tasks selected according to their behavioral profiles, promoting psychological recovery and sustained participation. Points, badges, and visual progress markers reinforce behavioral adaptation and awareness. Furthermore, environmental sensor data inform gamified ecobehavioral tasks (for example, promoting natural lighting or reducing HVAC usage), aligning with workplace sustainability goals outlined in EU Directive 2018/844. This enables the system to extend its motivational framework beyond individual well-being, fostering collective responsibility toward environmental performance. Given the sensitive nature of processing psychological and behavioral data, the system strictly complies with data protection regulations to ensure user privacy and ethical data handling.

4.1. Microservice-Based Architecture

The system follows a microservice architecture to ensure modularity, scalability, and fault isolation. Each core functionality is encapsulated within a dedicated service, deployed, and managed independently using Docker containers orchestrated in a cloud environment (e.g., AWS). The system architecture (see Figure 4) is structured around a set of modular components organized in a microservice-based framework, each responsible for a distinct functionality and implemented with appropriate technologies to ensure scalability, maintainability, and performance. The user interacts with the system through a lightweight web-based front-end developed using Flutter, which provides a seamless mobile experience. Biometric and behavioral telemetry from IoT devices is collected and pre-processed by a dedicated Data Collection Service, built in Python using Flask to expose efficient RESTful endpoints. The processed data is then forwarded to the AI Inference Service, which integrates a neural network model for stress-level classification and offers real-time predictions via REST APIs. Based on the inferred outputs and the user’s contextual history, the Survey Generation Service—developed using Spring Boot—dynamically produces personalized and context-aware questions, supporting versioning and user profiling. To enhance user engagement, a Gamification Service, built with Spring Boot and backed by MongoDB, manages the logic for rewards, point accumulation, badges, and user progression tracking. Security is enforced through an Authentication and Authorization module using Spring Security and JWT, which manages access control and protects user data in coordination with a centralized API Gateway. This gateway, implemented with WSO2, handles all routing, rate limiting, and logging, and changes in stress level are propagated asynchronously through Integromat across internal and external microservices. Deployment and container orchestration are achieved via Docker containers hosted on AWS, providing portability and elastic resource scaling. Data persistence relies on MongoDB, a NoSQL database selected for its flexibility in storing semi-structured information, such as biometric data, survey results, and gamification metrics, and for its horizontal scalability capabilities essential to system growth. All services communicate via RESTful APIs over HTTPS, with service discovery and orchestration managed by a lightweight control plane—events such as surveys, message queues, enabling loose coupling, and real-time reactivity. The architecture ensures that each module is independently deployable, testable, and replaceable, which supports continuous integration and rapid iteration. The use of open standards and interoperable protocols allows the architecture to evolve and scale seamlessly as new survey types or biometric metrics are introduced.

4.2. Use Case Simulation

In order to provide a concrete idea of how the system can operate within a real-world context, an application scenario involving an SS has been designed. The following section illustrates how the system components interact to enable seamless integration of data-driven stress detection, personalized feedback, and gamified behavior change within an industrial context. In a typical scenario, a knowledge worker wears a smartwatch and regularly uses a smartphone, continuous sources of multimodal data. Throughout the workweek, the system initiates uninterrupted acquisition of biometric, behavioral, and environmental data streams. Biometric indicators such as elevated resting heart rate during working hours, reduced sleep duration across multiple nights, and extended periods of sedentary behavior are collected via the smartwatch, while the smartphone captures behavioral patterns including post-work screen time and increased evening social media activity. In parallel, environmental sensor systems deployed in the workplace monitor contextual variables such as artificial lighting status (on/off), window opening, and HVAC system activity, providing insight into workspace conditions that may influence well-being and energy efficiency.

These heterogeneous data streams are transmitted to the AI Inference Service, where a neural network trained on multivariate temporal and contextual features performs advanced stress-level assessment. By jointly analyzing physiological responses, digital behavior, and environmental exposures, the system identifies subtle correlations—such as the relationship between inadequate lighting or poor ventilation and signs of cognitive fatigue or digital overstimulation—enabling a richer and more ecologically valid understanding of user stress. Upon detecting patterns indicative of rising mental fatigue or burnout risk, the system triggers a dynamic, context-aware SS: not in the form of traditional questions, but as proactive, gamified interventions. These include actionable suggestions such as opening a window, taking a short break, or reducing digital activity, aiming to support self-regulation and restore environmental and physiological balance. Although the present work relies on synthetic data for feasibility assessment, integrating both biometric and environmental variables as inputs to the neural network is supported by recent empirical research. For instance, Meng et al. [58] propose an AI-enhanced digital twin framework that fuses ECG-derived heart rate variability with environmental parameters such as temperature, humidity, and ventilation to accurately predict stress levels in real-time applications. Biometric data are acquired in parallel with environmental parameters such as temperature, humidity, and ventilation, using IoT-enabled sensors to ensure synchronized and continuous multimodal data collection. The physiological signals are segmented using dynamic sliding windows. HRV metrics are computed and input into a Random Forest classifier trained to predict stress levels across multiple categories. To improve transparency and model interpretability, SHAP (SHapley Additive exPlanations) values are calculated, allowing the identification of which specific features—both biometric and environmental—contribute most to each prediction. The framework also integrates a multi-scale intervention strategy, whereby predicted stress states are directly mapped to context-aware environmental adjustments, enabling real-time, actionable feedback at personal, spatial, or building levels. These findings support the multimodal input architecture adopted in this study, even though empirical validation in real-world settings remains a focus for future work. The Gamification Service activates a motivational loop by awarding points and proposing micro-challenges that encourage healthier digital habits and promote active recovery strategies. When the inference engine detects elevated stress levels, the user receives a real-time notification suggesting a short recovery break and an interactive mini-game, selected based on their motivational profile and designed according to Self-Determination Theory (SDT). These games, varying from mindfulness-based puzzles to goal-setting challenges, are crafted to enhance autonomy, competence, and relatedness—key psychological needs in gamification—while gently nudging behavior change. In addition to promoting individual well-being, the gamification engine is also informed by environmental sensor data. For instance, if sustained patterns of unnecessary artificial lighting or inefficient HVAC use are detected during occupancy periods, the system activates ecobehavioral challenges (e.g., “Open the window instead of using air conditioning”, or “Work with natural lighting for one hour today”). These tasks are aligned with sustainability goals outlined in the EU Directive 2018/844, which emphasizes the importance of improving energy performance and indoor environmental quality in buildings. The system fosters a culture of shared environmental responsibility across the workplace by translating energy-efficient behaviors into game dynamics and rewards. Each completed challenge yields additional points and unlocks visual badges that mark progress. A cumulative score is maintained in the user’s profile and displayed on a personalized dashboard, promoting long-term motivation through both intrinsic and extrinsic rewards. This scenario emphasizes how gamification is not peripheral but essential in transforming passive monitoring into active behavior change, closing the loop between awareness, action, and well-being.

The gamification layer is conceptually grounded in Self-Determination Theory (SDT), a psychological model that explains human motivation through satisfying three innate psychological needs: autonomy, competence, and relatedness. In digital health and behavioral interventions, SDT has been extensively applied to account for sustained engagement and internalization of health-promoting behaviors.

Within this framework, the SS system supports autonomy by offering users personalized, context-aware questions that reflect their physiological and behavioral data. It allows them to choose how and when to engage with the feedback, reinforcing a sense of control and agency. Competence is promoted through challenges and reward mechanisms that provide users with tangible progress indicators, such as improved sleep patterns or reduced digital overload; visual feedback and gamified milestones further strengthen users’ perception of efficacy. Although designed primarily for individual use, the platform addresses relatedness by optionally incorporating shared badges or leaderboard features, fostering a sense of community and shared purpose among users facing similar well-being challenges. By aligning its design with the principles of SDT, the system goes beyond short-term behavioral compliance, aiming to cultivate long-term intrinsic motivation for healthier and more sustainable habits.

5. Results

This study introduces and evaluates an intelligent system that transforms the traditional concept of SSs in industrial environments. The proposed solution goes beyond conventional survey-based mechanisms, typically based on predefined questions and periodic assessments, leveraging real-time sensor data, AI-driven inference, and personalized context-aware interventions. Rather than passively collecting responses, the system is an adaptive, proactive component within workplace ecosystems. It interprets biometrical and environmental signals, detects deviations from normative baselines, and provides timely feedback to maintain user well-being and operational performance. This approach offers an interesting paradigm for behavioral data acquisition and feedback, where measurement is seamlessly embedded into routine activities and supports self-regulation without disruption. This redefinition addresses key limitations observed in the literature: traditional survey tools often interrupt workflows, lack contextual relevance, and fail to capture dynamic variations in well-being. In contrast, the proposed system integrates IoT-based sensing with AI inference to provide adaptive, personalized, and embedded feedback. The system architecture is modular and based on microservices, enabling decentralized data acquisition, ethical governance, and seamless integration into existing infrastructures. Importantly, the motivational dimension is addressed through gamification strategies, which increase user engagement and reduce resistance to behavioral feedback mechanisms. What emerges from this work is not merely a refinement of survey content, but the creation of an interaction model in which the survey becomes a reflective and reactive stimulus, triggered by events and embedded within a broader feedback loop based on behavioral inference and adaptive intervention. The system transforms the SS from a passive measurement tool into an active and responsive component of workplace dynamics, capable of promoting well-being, self-regulation, and sustainable behavior without disrupting operational flows. The innovative contribution of this research unfolds along four principal axes. First, from a process perspective, the system introduces a closed-loop, event-driven model in which biometric and behavioral signals—such as stress indicators, sedentary behavior, and environmental factors—are continuously monitored through IoT devices. When stress or overload patterns are detected by the AI-based inference module, the system does not respond with a traditional questionnaire, but with a context-aware behavioral challenge, reinforcing its design principle of low intrusiveness and passive monitoring. Second, in terms of interaction, these stimuli are not static forms but elements of a dynamic behavioral dialogue. They are generated in real time based on the user’s psychophysiological state and followed by gamified feedback, thus transforming the interaction from a monitoring action into a proactive behavioral catalyst fully embedded in the daily workflow. Third, from an architectural standpoint, the process is enabled by a modular, microservice-based infrastructure that decouples sensing, inference, intervention, and user engagement. This allows for real-time scalability, domain adaptability, and ethical compliance through robust data governance mechanisms such as data minimization, secure APIs, and consent-based logic. Finally, from a motivational angle, the integration of gamification plays a pivotal role. By incorporating playful and interactive mechanisms, the system enhances user participation and commitment, strengthening engagement and digital adoption while overcoming cultural resistance. A synthetic use case was developed to demonstrate the feasibility of this approach. The scenario involves a knowledge worker in a mid-sized industrial organization, monitored over five days via wearable and environmental sensors. Biometric data such as heart rate variability (HRV) and sleep duration are inferred from smartwatch data, while environmental conditions—such as artificial lighting, HVAC status, and window position—are monitored through ambient IoT infrastructure. So, a time-based simulation was conducted over a five-day period. This multi-day horizon was chosen to replicate a typical working week and to allow for the observation of biometrical and environmental data over time. Monitoring over several days is essential to establishing a contextual baseline, detecting deviations from standard patterns, and assessing how sustained or cumulative stress conditions might evolve and be recognized by the system. The simulation assumes the presence of continuous data streams from both wearable and ambient IoT sensors. Biometric signals such as HRV and sleep duration are generated synthetically to reflect variations that might occur in a real user. For example, HRV values are simulated using RR intervals (time between two consecutive heartbeats, measured from peaks in the ECG signal) with decreasing variability over time, while sleep duration is modeled based on typical adult sleep patterns affected by workload. In parallel, environmental data such as lighting conditions, HVAC system status, and window position are set to represent environmental states commonly found in office spaces. On Day 3 of the simulation, a specific set of conditions was modeled to represent a ’triggering event’. This day was characterized by a marked drop in HRV (below 30 milliseconds), reduced sleep duration (less than 5 h), and a fully closed, artificially conditioned work environment: artificial lighting was active, HVAC was running, and windows were closed. These values represent critical thresholds identified in the system’s inference logic as indicators of possible cognitive overload or psychophysiological fatigue. The co-occurrence of these parameters across multiple modalities—biometric, behavioral, and environmental—activated the AI inference module. The system classified the user’s condition as a high-stress state and triggered an intervention phase. Instead of initiating a survey or manual query, the system autonomously generated a context-aware, gamified recommendation. This took the form of a real-time notification delivered via the user interface, encouraging the user to restore environmental balance and earn a motivational badge by opening a window and turning off HVAC for a defined period. This type of feedback is designed not only to alleviate environmental stressors but also to reinforce proactive behaviors through a reward-based mechanism, all without interrupting the user’s workflow.

This process illustrates how the system can detect early warning signs of fatigue or stress and autonomously generate personalized interventions based on physiological and environmental parameters. The synthetic input data used to trigger the feedback loop are shown in

Table 4. Synthetic biometric and environmental data used in the simulation.

Day	HRV (ms)	Sleep (h)	Light (ON)	HVAC (ON)	Window (Closed)	Trigger
Day 1	52	7.2	No	No	No	No
Day 2	48	6.8	Yes	Yes	Yes	No
Day 3	20	4.9	Yes	Yes	Yes	Yes
Day 4	36	6.3	Yes	No	No	No
Day 5	45	7.0	No	No	No	No

.

Figure 5 shows the time-series data, the system logic pipeline from sensing to inference and response, and the resulting user interface interaction. The figure includes two simulated time-series curves across five consecutive days. The black curve represents heart rate variability (HRV), calculated using the RMSSD (Root Mean Square of Successive Differences) metric. The RMSSD is a widely adopted index in the domain of short-term HRV analysis and is sensitive to parasympathetic activity. The HRV values were generated synthetically to reflect plausible physiological variations in response to increasing cognitive load and environmental discomfort, with an apparent decline visible leading up to Day 3. The red curve displays sleep duration (in hours), also modeled synthetically based on fatigue accumulation typically observed in industrial knowledge workers. The sleep trend mirrors the HRV decline, reaching a low point of under 5 h on Day 3. This co-occurrence of sub-threshold values—HRV below 30 ms and sleep below five hours—constitutes the critical triggering condition used by the system’s AI module. The crossing of these thresholds is annotated in the graph as the “triggering point,” emphasizing how multimodal data are jointly analyzed to identify latent overload conditions. This fusion of biometric indicators, rather than relying on single-variable heuristics, enhances system reliability and reduces false positives due to transient anomalies. By using time-based simulation across multiple days, the figure illustrates how the system builds a temporal context to evaluate user states dynamically. The pipeline diagram and the mobile interface mockup complement this process, showing how sensed data is processed into actionable, gamified feedback in a real-time, non-intrusive manner.

This redefinition—enabled by the synergy between IoT sensing, real-time AI, gamification, and motivational design—represents the core scientific and technological innovation of this work. It enables the development of a new generation of intelligent systems in organizational contexts. Here, data collection is no longer separate from intervention, and employees are not merely subjects of measurement but active agents of engagement.

6. Discussion

The analysis of current scientific literature reveals that SS applications remain limited in quantity and scope. Although there is growing interest in intelligent data collection systems, a significant gap still exists in applying such solutions in industrial and organizational contexts—particularly about real-time monitoring of worker well-being. This work contributes to closing that gap by proposing a conceptual model that integrates sensing, inference, and gamified feedback, all without requiring explicit user input. The simulated case study presented in this paper highlights how the proposed system can transition from passive monitoring to active behavioral intervention. The system delivers timely, context-sensitive, and actionable feedback by leveraging a synergistic combination of biometric and environmental data, real-time AI processing, and motivational design principles. This approach is aligned with behavioral science literature, which emphasizes the importance of immediacy and contextual relevance in stimulating sustainable behavioral change in the workplace. A concrete use case was designed to illustrate the architectural and functional potential of the system. Although conceptual, this simulation reveals both the technical feasibility and the operational value of the system, providing a foundation for future empirical validation. Upcoming work will focus on piloting the architecture in real organizations, evaluating key performance metrics such as user engagement, behavioral compliance, and the longitudinal impact on well-being and productivity. Despite these promising directions, several challenges remain. Among the most critical issues are privacy, user autonomy, and generalizability. Integrating physiological and behavioral monitoring in the workplace naturally raises ethical and legal concerns that must be addressed from the early stages of system design. Ensuring a high level of perceived security among users is essential for system acceptance, alongside full compliance with international regulations such as the General Data Protection Regulation (GDPR) and national data protection laws. To meet these requirements, the development and deployment of SS systems must be explicitly guided by the principles of privacy-by-design [59]. These include proactivity and prevention, privacy by default, data protection integration into system architecture, full functionality without compromising confidentiality, end-to-end security, transparency and accountability of data processes, and user-centricity in every interaction phase. The proposed architecture is designed to serve as a reference model in this context. It incorporates key elements that enforce data security and ethical processing. Security and access control are managed through the implementation of authentication and authorization mechanisms using Spring Security and JSON Web Tokens (JWT), which support decentralized identity verification and access management. The use of a centralized WSO2 API gateway ensures secure routing, throttling, and monitoring of inter-service communications, effectively preventing unauthorized access. A microservice architecture is adopted to reduce the centralization of sensitive data, minimize the system’s attack surface, and improve modularity and resilience. Scalability and portability are ensured through the use of Docker containers, which allow for efficient and secure deployment across heterogeneous environments. Finally, data management is handled using MongoDB, which provides flexible storage of semi-structured biometric and behavioral data, along with built-in horizontal scalability and encryption capabilities. Ultimately, a strategic trade-off between protecting personal data and the granularity and utility of the behavioral insights produced must be identified. Achieving this balance is critical for the responsible, ethical, and scalable implementation of SS systems. Such tools can only gain user trust and provide real value in modern industrial and institutional ecosystems by reconciling technological potential with ethical responsibility.

7. Limitations

From a methodological perspective, the research results have some limitations. First, the relative novelty of the Smart Survey (SS) topic means there is still a limited number of high-quality studies available, which may reduce the generalizability of the findings. Another limitation concerns the inclusion/exclusion criteria: the analysis focused solely on articles explicitly mentioning “Smart Survey,” potentially overlooking relevant studies using different terminology. Moreover, IoT technologies’ diversity and practical applications complicate direct and standardized comparisons across studies. Some studies may have also been excluded due to language barriers or limited access to full-text articles. The scarcity of literature specifically addressing IoT solutions in the industrial sector further constrains the completeness of the collected evidence. However, this gap highlights a promising direction for future research, especially regarding integrating innovative survey tools with IoT solutions to optimize data collection, analysis, and utilization in industrial and business contexts. A simulation was performed to validate the theoretical soundness of the proposed solution. Nonetheless, this approach has inherent limitations regarding theoretical design validation. The simulation used synthetic data to assess the system’s potential experimentally. This methodology was chosen to enable an initial evaluation within a controlled and reproducible environment during the early stages of the study. While the preliminary results are encouraging, future work will further focus on deploying the system in real industrial settings to substantiate the model’s validity under operational conditions. Empirical usability evaluation, involving real users interacting with the system in realistic scenarios, is essential to identifying usability issues and assessing acceptance and effectiveness in practical contexts. Currently, the study is limited by the absence of such empirical evaluations of the prototype. Therefore, future research should include user-centered empirical studies to complement theoretical validation, ensuring the solution is functional and user-friendly in real-world applications.

Limitations of Synthetic Data

Synthetic data are increasingly valued for their potential across various application domains [60], especially in contexts characterized by data scarcity or difficulty in data collection. However, they present significant limitations, such as the lack of semantic depth and linguistic richness typical of real data, which reduces their effectiveness in classification tasks [61].

In this preliminary study, we used synthetic data to contribute to the literature on Smart Surveys, a still underexplored field, and to validate an architectural solution applicable in industrial contexts. The simulation with synthetic data confirmed the system’s ability to process real-time inputs to detect stress and trigger personalized interventions.

In future developments, it will be essential to compare synthetic data with real data and improve strategies for evaluating their reliability.

8. Conclusions

Reviews of the literature play an essential role as the basis for all types of research. For this reason, the study conducted not only contributes to enriching a still limited body of literature but also provides a solid basis for knowledge development and guides future scientific work. It identifies their potential uses and outlines key challenges. It highlights how IoT technologies—traditionally employed in industrial environments for predictive analytics—can also facilitate collecting and statistically analyzing environmental data to promote workers’ well-being and engagement. This article presents a framework for implementing SSs in industrial settings to collect information on employee behaviors related to energy consumption and workplace well-being. The system leverages low-cost IoT devices to acquire relevant data, enabling continuous and intelligent monitoring of the work environment. To encourage widespread adoption and usage of such solutions, a gamified application has been developed that simulates user behavior through game dynamics. The goal is to increase user engagement, make the experience more interactive and stimulating, and promote conscious use of the system to improve daily habits related to energy efficiency and personal well-being. Our approach to integrating gamification techniques with SSs in industrial environments arises from the imperative to address cultural barriers that hinder their widespread adoption. To this end, we designed a modular and scalable system architecture that integrates SSs with gamification mechanisms, making the interaction more natural, engaging, and sustainable over time. To demonstrate the feasibility of our approach, a simulation was conducted replicating a typical five-day work scenario with synthetically generated biometric and environmental data. The simulation validates the system’s ability to process real-time inputs (e.g., HRV, sleep, HVAC usage) to detect stress conditions and trigger personalized interventions. This step provides empirical support for the robustness and applicability of the proposed framework in realistic industrial settings. The next step will involve comparing two configurations of SSs—one enhanced with gamification elements and one standard—to assess their impact on engagement and usability and verify whether the introduction of game-based components fosters more active participation and higher user satisfaction. Finally, this study opens new research perspectives on the joint use of SSs and gamification in industrial contexts, emphasizing the need to promote an organizational culture that is open to change. Only through such advancements will it be possible to fully exploit the potential of SSs while addressing current technological limitations in its design and implementation. Future work will include developing and implementing a pilot project to apply the proposed solutions within a real operational context. An industrial pilot has already been identified and will represent a crucial phase for the empirical validation of the platform, enabling the collection of real-world data to be directly compared with the generated synthetic data. The timeline for the pilot’s completion will depend on the operational challenges encountered during its execution.