Overcoming Barriers to Digital Transformation towards Greener Supply Chains in Automotive Paint Shop Operations

Abstract

Given the resource-intensive nature of automotive manufacturing processes and their potential to substantially contribute to ecological footprints, the integration of sustainable logistic practices in the context of digital transformation becomes imperative. This paper focuses on the implementation of green supply chain strategies within the automotive sector, targeting significant risks associated with environmental impact, specifically in the critical domain of automotive paint shops. Automotive paint shops indeed play a significant part in determining the overall sustainability of automotive production. Recognized for their role in vehicle esthetics and corrosion protection, the sustainable integration of these facilities is crucial in the pursuit of a greener automotive future. A comprehensive multi-criteria decision-making framework is herein proposed as a valuable tool in pinpointing the most critical barriers to digital transformation and simultaneously prioritizing suitable green logistic strategies in the context of automotive paint shop risk-management procedures. The practical utility of the model extends to practitioners in the automotive paint shop supply chain, particularly those engaged in digitalizing critical operations, facilitating well-informed decision-making aligned with environmental sustainability goals. The findings of this research highlight the critical importance of implementing tailored strategies, including crisis preparedness, transparent communication, proactive outreach, and strategic investments in technology and partnerships, to address barriers and enhance sustainability practices within automotive paint shop operations, thereby contributing to the overall resilience and long-term viability of automotive supply chains.

1. Introduction and Objectives

The automotive industry is a major player in the global economy, but concerns about its environmental impact have grown [1]. Efforts to integrate sustainability face notable challenges, particularly within automotive paint shops, essential facilities in the manufacturing process. These establishments often encounter obstacles in implementing eco-friendly practices [2]. Of particular concern is the suspected pollution surrounding painting booths, emphasizing the critical need for sustainable measures.

Based on previous research [3], it is evident that this issue presents a substantial risk that requires immediate attention. Addressing the associated challenges is crucial for advancing green supply chain operations and mitigating the potential impact of suspected pollution near painting booths. Implementing a proactive environmental monitoring program in and around automotive paint shops emerges as a potential solution. Such a program would regularly assess air and water quality as well as soil conditions, aiming to detect any potential pollution stemming from painting booths. By promptly reporting abnormal findings, the automotive industry could contribute to minimizing its environmental footprint and promoting greener manufacturing processes [4].

However, the implementation of such a proactive program in automotive paint shops may encounter barriers on multiple fronts [5]. Financial constraints, technological adaptations, regulatory compliance, data management complexities, cultural resistance, collaboration challenges, and the risk of greenwashing [6] further complicate the effective establishment of these programs. Navigating these diverse barriers requires a comprehensive strategy addressing financial, technological, regulatory, cultural, and collaborative aspects to ensure the success and sustainability of environmental monitoring initiatives in the automotive sector. The main research hypothesis is that the presence of the mentioned barriers significantly impedes the establishment of effective monitoring plans in automotive paint shops. Furthermore, it is hypothesized that analyzing these barriers can yield valuable insights, informing the development of strategies aimed at improving the performance of monitoring plans in these settings. To achieve such an aim, this paper addresses this problem through a multi-criteria decision-making (MCDM) strategy, allowing for a systematic evaluation of barriers and the prioritization of management solutions. By considering several criteria, the most viable and comprehensive strategies can be selected to overcome obstacles and ensure the successful implementation of environmental monitoring initiatives in the automotive industry.

This paper aims to thoroughly analyze the primary barriers associated with implementing monitoring programs to mitigate the risk of suspected pollution in automotive paint shops. By meticulously reviewing the existing literature, the research aims to identify and formalize a comprehensive set of obstacles within this sector. Subsequently, the study proposes an integrated MCDM approach. Initially, barriers will be categorized based on their management complexity using the ELimination Et Choix Traduisant la REalité (ELECTRE) TRI method. Following this categorization, barriers will undergo further evaluation within each category using the technique VIsekriterijumsko KOmpromisno Rangiranje (VIKOR), which is the Serbian name standing for multi-criteria optimization and compromise solution. This dual-step process is designed to identify and prioritize barriers, providing a targeted approach for defining effective green strategies aimed at overcoming these identified obstacles in the automotive paint shop sector. The integration of ELECTRE and VIKOR methods in prior literature [7] has proven to be successful in addressing complex decision-making challenges. However, to the best of the author’s knowledge, the specific utilization of such a framework to address the complexities inherent in the automotive paint shop sector with a specific focus on sustainability represents an innovative approach. This study proposes to fill this gap in the literature by proposing the application of ELECTRE and VIKOR methods within the automotive paint shop domain and is specifically tailored to address the challenges of implementing monitoring programs and promoting sustainable practices. By adapting and applying these established methodologies to this unique problem, this research offers insights and solutions that can contribute to improved environmental performance and operational efficiency within automotive paint shops.

The structure of this paper is outlined as follows. Section 2 presents a comprehensive literature review, culminating in the formalization of barriers, as specified earlier. In Section 3, the rationale for selecting the previously mentioned MCDM techniques is explained, accompanied by pertinent methodological details. Section 4 develops a case study focused on the Italian automotive paint shop previously analyzed in [3]. Section 5 reports a detailed discussion about managerial insights. The concluding remarks are discussed in Section 6.

2. Literature Review

2.1. Environmental Impact and Technological Innovations in Automotive Paint Shops

The automotive industry, a cornerstone of global manufacturing, stands as a critical contributor to economic growth. An empirical analysis of industrial policies [8] revealed a tendency to ecologically modernize rather than deeply transform the automotive sector, emphasizing the broader need for transformative approaches to address environmental concerns in the transportation landscape. However, this transformation is accompanied by an undeniable environmental impact, notably within automotive paint shops. The environmental impact of automotive paint shops is indeed closely linked to energy-intensive operations, as discussed by Giampieri et al. [9]. The authors highlight the significance of air management systems within paint booths, emphasizing their role in maintaining optimal temperature and humidity to prevent paint defects. The environmental repercussions of automotive paint shops stem from the release of volatile organic compounds (VOCs) [10], hazardous air pollutants (HAPs) [4], and other noxious emissions associated with the painting process [11]. As regulatory frameworks tighten and environmental consciousness rises, there is a growing need to scrutinize the technologies employed in these facilities.

Addressing this need, a methodology for data-driven energy demand prediction and performance bench-marking was introduced in [12] to systematically analyze and derive improvement strategies for reducing industrial greenhouse gas emissions and was validated within the automotive industry. Furthermore, the painting technologies employed in automotive paint shops play a central role in the environmental equation. With this perspective, Cavalcante et al. [13] introduce a neural network predictive control (NNPC) system for optimizing temperature control in an automotive oven’s drying and curing processes, effectively reducing costs while maintaining painting quality.

Traditional painting methods often involve the use of solvent-based coatings, which pose a substantial risk of pollution due to the release of harmful chemicals into the atmosphere. Given automotive paint’s dual role in esthetics and environmental protection, it is essential to understand its composition, which includes binders, solvents, pigments, and additives. As observed in [14], the controlled application process adheres to strict specifications for film thickness, chemical composition, and post-processing steps, such as baking to enhance durability. The emphasis on pollution risk arises from the use of solvent-based coatings. To address this, there is a growing interest in exploring eco-friendly alternatives, like water-based paints, which aim to minimize the emission of harmful chemicals during the automotive painting process, fostering a more sustainable and environmentally conscious approach [15]. The shift towards water-based and powder coating technologies has presented itself as a more environmentally friendly alternative. However, even these alternatives are not without environmental concerns, as they may still involve the use of certain chemicals with potential ecological consequences [16].

2.2. Integration of Monitoring Systems for Environmental Compliance

In response to the escalating environmental concerns associated with automotive paint shops, there is a growing emphasis on adopting technologies and strategies that mitigate pollution and health risks [17]. Monitoring systems have emerged as a critical component in this endeavor. This proactive approach aims to enhance air quality within paint shop environments, acknowledging the critical role of monitoring systems in containing and addressing potential pollution sources, as highlighted in recent studies [18]. Real-time monitoring systems, when integrated into the operational framework of paint shops, provide continuous oversight of emissions and other environmental parameters. This enables the prompt identification of deviations from permissible environmental standards, facilitating timely corrective actions and minimizing the environmental impact of these facilities. Dacal-Nieto et al. [19] explore the background and rationale for industrializing a data analytics system in manufacturing, emphasizing its relevance in automotive paint shops. By prioritizing quality enhancement, the optimization of processes, and the prompt identification of parameter deviations, their research illustrates the economic benefits at the initial stages. This integration is further exemplified in automated automotive paint shops, where a sophisticated network of sensors and controllers continuously monitors the coating process for vehicle bodies-in-white. These data are stored in the Painted Surface Performance Management (PSPM) system, and machine learning models identify key features influencing painted vehicle outcomes. These features contribute to improved operational first time quality (FTQ). Using data from 2020–2021 of an automotive paint shop, the study [20] highlights the role of monitoring systems in optimizing processes and offers insights for improvement.

Despite the undeniable benefits of monitoring systems, their implementation is not without challenges. Financial constraints [21], technological complexities [22], and resistance within organizational structures are among the prominent barriers faced in adopting these systems [23]. The financial implications of investing in monitoring technologies, coupled with uncertainties regarding the return on investment, often deter organizations from taking proactive measures [24,25,26]. Additionally, the integration of sophisticated monitoring systems may require substantial technological upgrades [27], posing challenges for facilities with outdated infrastructure. Overcoming these barriers represents a complex decision-making problem where various criteria come into play, expressing the need for a multidimensional approach. In this context, MCDM methods represent particularly useful methodological frameworks. These methods provide a systematic and structured way to evaluate and prioritize various criteria, enabling decision-makers to weigh different factors, assess trade-offs, and ultimately make informed and robust decisions. By adopting MCDM methods [28], decision-makers can effectively address the complexities inherent in decision-making processes, leading to more thoughtful, well-informed, and well-balanced outcomes.

2.3. Multi-Criteria Decision-Making in Overcoming Implementation Barriers

MCDM methods have been extensively applied to solve real problems in the automotive sector. Garg et al. [29] propose an integrated fuzzy group MCDM framework dealing with the complexities of industrial robot selection, especially with the addition of diverse specifications from manufacturers. Ma and Li [30] introduce a decision support system tailored for assessing the quality of suppliers in manufacturing firms. The system’s effectiveness is exemplified through a case study conducted at a major automotive corporation, illustrating its strength in comparison with conventional MCDM techniques. Suraraksa and Shin [31] analyze multiple criteria for selecting suppliers as well as monitoring their efficacy in the context of the automotive industry. Rodrigues et al. [32] analyze key performance indicators (KPIs) that are essential for proficient performance monitoring in the dynamic landscape of rapid technological advancements. Conducting a case study within the automotive press molding sector, the authors pinpoint and rank nine KPIs in alignment with balanced scorecard criteria. Zhou et al. [33] select the optimal automotive component for quality enhancement using data from post-sale channels. The author proposes a novel hybrid MCDM framework that not only enhances decision accuracy over conventional methods but also demonstrates adaptability in managing quality improvement practice (QIP) decision-making.

Despite the widespread use of MCDM in the literature [34,35,36], a gap in its application is identified in addressing barriers related to implementing monitoring systems for evaluating and managing the risk of pollution in automotive paint shops.

Table 1. Main categories, barriers that emerged from the literature, and related descriptions.

Category	Barrier	Description
Financial Barrier (FB)	FB1 Limited Budget [24,25,26]	• The cost of implementing and maintaining an environmental monitoring program may strain the available budget, impacting the scope and effectiveness of the program.
	FB2 High Initial Investment [25,26]	• Purchasing monitoring equipment and establishing a comprehensive program may require a significant upfront investment.
	FB3 Operational Costs [37,38]	• Ongoing operational expenses, such as maintenance, calibration, and data analysis, may strain financial resources.
	FB4 Cost of Expertise [25]	• Hiring or training personnel with the required expertise to operate and interpret monitoring equipment can be expensive.
	FB5 Unforeseen Expenses [39]	• Unexpected costs, such as equipment breakdowns or the need for emergency response, can strain financial resources.
Technological Challenge (TC)	TC1 Complex Equipment [24]	• Monitoring equipment may be complex, requiring skilled operators and periodic maintenance for accurate results.
	TC2 Integration Issues [40]	• Integrating different monitoring tools and systems into a seamless program may pose technical challenges.
	TC3 Data Management [40]	• Handling and interpreting the large volumes of data generated by monitoring activities can be overwhelming.
	TC4 Technological Obsolescence [41]	• Rapid advancements in monitoring technology may render existing equipment obsolete, requiring frequent updates.
	TC5 Training Requirements [42]	• Adequately training personnel to use and troubleshoot monitoring equipment may be time-consuming and technically demanding.
Regulatory Compliance (RC)	RC1 Changing Regulations [43]	• Adapting the monitoring program to evolving environmental regulations and standards can be challenging.
	RC2 Legal Complexity [44]	• Navigating the legal framework associated with environmental monitoring and reporting may be complex.
	RC3 Documentation Requirements [44]	• Ensuring accurate and timely documentation to meet regulatory standards can be burdensome.
	RC4 Penalties for Non-Compliance [45]	• Fear of penalties for non-compliance with regulations may hinder the implementation of the monitoring program.
	RC5 Conflict of Standards [45]	• Balancing and aligning monitoring practices with multiple regulatory standards may be difficult.
Human Resource Constraint (HRC)	HRC1 Expertise Shortage [25,44,46]	• Finding and retaining skilled personnel with expertise in environmental monitoring may be challenging.
	HRC2 Workforce Training [47]	• Training existing staff in monitoring procedures and protocols may require time and resources.
	HRC3 Employee Resistance [25,44]	• Employees may resist changes in their work routines or additional responsibilities associated with the monitoring program.
	HRC4 Staff Turnover [48,49]	• High turnover rates can lead to a loss of institutional knowledge and continuity in the monitoring program.
	HRC5 Competing Priorities [42]	• Existing job responsibilities may compete with the time and attention required for effective monitoring.
Communication and Reporting (CR)	CR1 Data Interpretation [50]	• Communicating complex environmental data to non-specialists may pose challenges.
	CR2 Timely Reporting [50]	• Ensuring prompt reporting of abnormal findings to relevant authorities may be hindered by bureaucratic processes.
	CR3 External Communication [50]	• Effectively communicating monitoring results to the public, neighboring communities, or stakeholders may be challenging.
	CR4 Language Barriers [44]	• Dealing with language diversity among stakeholders and authorities may impede effective communication.
	CR5 Stakeholder Engagement [44,51]	• Involving and engaging stakeholders in the monitoring process may require significant effort.
Public Relations and Reputation Management (PRRM)	PRRM1 Negative Public Perception [52]	• The mere existence of an environmental monitoring program may be perceived negatively by the public.
	PRRM2 Reputation Risks [52]	• Abnormal findings or incidents may lead to damage to the company’s reputation.
	PRRM3 Media Scrutiny [49]	• Media attention and coverage may amplify the impact of environmental issues, affecting public perception.
	PRRM4 Competitive Disadvantage [49]	• Publicizing monitoring efforts may be seen as a weakness by competitors.
	PRRM5 Community Relations [49]	• Balancing the company’s interests with those of the local community may be challenging, impacting public relations.

reports a set of significant barriers related to the implementation of monitoring systems in the automotive sector that emerged from the analysis of the literature. The identified barriers have been described and classified into six categories, namely Financial Barrier (FB), Technological Challenge (TC), Regulatory Compliance (RC), Human Resource Constraint (HRC), Communication and Reporting (CR), and Public Relations and Reputation Management (PRRM). Addressing these barriers within each category is essential for the successful implementation and sustained effectiveness of proactive environmental monitoring programs. Recognizing the significant flexibility of MCDM techniques, the present paper introduces an integrated MCDM approach to fill the previously mentioned research gap, providing a detailed justification for the chosen methodology in the next section.

3. Methodological Approach

The ELECTRE TRI and VIKOR methods are herein proposed as an integrated MCDM approach to initially classify barriers associated with the implementation of monitoring programs for suspected polluted areas in automotive paint shops into distinct complexity classes, and to subsequently evaluate the barriers within each identified class. Such a methodological combination provides a balanced approach that considers both outranking relationships and compromise solutions. ELECTRE TRI is particularly useful to handle partial or imprecise information, while VIKOR excels at identifying compromise solutions when faced with multiple criteria. This combination offers a more nuanced and adaptable approach compared to other MCDM methods, making it well-suited to deal with the complexity of the decision-making problem faced. The advantages of implementing such a MCDM integration are herein discussed, with methodological details reported in the following subsections.

• ELECTRE TRI and VIKOR can offer a robust framework for categorizing barriers based on their management complexity. The two-step process allows for a comprehensive classification of barriers into distinct classes, providing a nuanced understanding of the varying levels of complexity associated with each barrier.
• VIKOR’s ability to evaluate alternatives [53] makes it particularly suitable for prioritizing barriers within each complexity class. This ensures that attention is directed towards the most critical obstacles first, facilitating a more targeted and efficient approach to addressing barriers in the implementation of monitoring programs.
• Both ELECTRE TRI and VIKOR are capable of handling multiple criteria simultaneously [54]. This is crucial in the context of monitoring program implementation, as barriers often involve various factors. The methods enable a holistic assessment that considers diverse aspects, providing a more complete picture of the barriers.
• ELECTRE TRI and VIKOR are known for their ability to handle situations with incomplete or imprecise information [55,56]. In real-world scenarios, obtaining precise data for all criteria might be challenging. These methods allow for a certain degree of flexibility and adaptability when dealing with imperfect information, enhancing their practical applicability.
• The combination of ELECTRE TRI and VIKOR is adept at handling trade-offs between conflicting criteria [57,58]. In the context of barrier categorization and evaluation, there are often trade-offs between different aspects of complexity. These methods are effective in finding a balance, ensuring that the chosen barriers for attention represent a well-considered compromise among multiple criteria.

The proposed methodological approach, integrating ELECTRE TRI and VIKOR, will be practically applied in Section 4 to provide support for an Italian automotive paint shop, facilitating informed decision-making processes and enhancing operational efficiency in overcoming the most significant barriers to digital transformation.

3.1. ELECTRE TRI Sorting Barriers

The implementation of ELECTRE TRI involves two main phases conducted sequentially. Initially, outranking relations are established based on principles of concordance and discordance. Subsequently, these specified relations are applied to sort alternatives into classes, reflecting their shared features. This categorization process can be executed through two distinct procedures. The following input information needs to be collected before initiating the application: evaluation criteria $B_k$, under which alternatives are assessed; criteria weights $w_k$, indicating the relative importance of criteria; reference profiles $P_j$, characterized by specific evaluations under each criterion and defined by two limits $p_h$ and $p_{h+1}$; classes $C_h$, identified by reference profiles; alternatives $A_i$, with corresponding evaluations $B_k\left(A_i\right)$ under each criterion; threshold value $\lambda$, known as the “cutting level” and comprised between 0.5 and 1, that is required to complete the initial stage of ELECTRE TRI. Additionally, indifference, strong preference, and veto thresholds, respectively denoted as $I_k$, $S_k$, and $V_k$, have to be established in association with the outranking relations. $I_k$ signifies the minimum difference necessary to establish a preference between a pair of elements, $S_k$ denotes the minimum difference required to indicate a strong preference, and $V_k$ indicates the minimum difference signaling incompatibility.

The first stage entails establishing an outranking relation by assessing each alternative in comparison to the class boundaries, particularly with the reference profiles. The subsequent primary steps must be executed in a sequential manner.

• To calculate partial concordance indices for each criterion, pairwise comparisons are conducted between each alternative $A_i$ and the designated reference profiles $P_j$. Subsequently, the concordance indices, labeled as $C_k(A_i,P_j)$, are computed for each criterion $B_k$ using Formula (1).

  $$C_k(A_i,P_j)=\left\{\begin{array}{c}1if[B_k\left(P_j\right)-B_k\left(A_i\right)]\le I_k\hfill \\ 0if[B_k\left(P_j\right)-B_k\left(A_i\right)]\ge S_k.\hfill \\ \frac{B_k\left(A_i\right)-B_k\left(P_j\right)+S_k}{S_k-I_k}\mathrm{otherwise}\hfill \end{array}\right.$$

  (1)
  The aggregated concordance index $C(A_i,P_j)$ is then calculated by consolidating and applying weights to the concordance indices for each criterion in the following manner:

  $$C(A_i,P_j)=\frac{{\sum }_{k=1}^K{w}_k·C_k(A_i,P_j)}{{\sum }_{k=1}^K{w}_k}.$$

  (2)
• Calculating partial discordance indices for each criterion is achieved by utilizing Formula (3).

  $$D_k(A_i,P_j)=\left\{\begin{array}{c}1if[B_k\left(P_j\right)-B_k\left(A_i\right)]>V_k\hfill \\ 0if[B_k\left(P_j\right)-B_k\left(A_i\right)]\le S_k.\hfill \\ \frac{B_k\left(P_j\right)-B_k\left(A_i\right)-S_k}{V_k-S_k}\mathrm{otherwise}\hfill \end{array}\right.$$

  (3)
• Calculating outranking credibility indices is accomplished using Formula (4).

  $$\delta (A_i,P_j)=C(A_i,P_j)·\frac{{\prod }_{k\in K *}(1-D_k(A_i,P_j))}{1-C(A_i,P_j)}.$$

  (4)
  Consider $K^{*}$ as the subset of criteria where $D_k(A_i,P_j)>C(A_i,P_j)$. When there is no veto threshold, the credibility index $\delta (A_i,P_j)$ equals the aggregated concordance index $C(A_i,P_j)$.
• Applying the particular form of outranking relation requires the use of the cutting level $\lambda$. Essentially, $\lambda$ acts as the threshold for $\delta (A_i,P_j)$ to support the hypothesis that $A_i$ outranks $P_j$. This value, $\lambda$, falls within the range of 0.5 to 1, and it must exceed the value equal to $1-(\mathit{highestweight}/\mathit{totalweight})$.

The subsequent phase of ELECTRE TRI comprises the classification of alternatives into classes using two potential approaches: the pessimistic and the optimistic rules. In most cases, the pessimistic procedure is favored over the optimistic approach. This preference is due to the pessimistic method’s tendency to allocate alternatives to classes characterized by lower profiles, thereby yielding more conservative outcomes.

In adherence to the pessimistic (or conjunctive) approach, an alternative $A_i$ is sorted to class $C_h$ if the condition $A_i$S$P_j$ holds true. The procedure is implemented through two steps: (1) systematically comparing each alternative with class boundaries, where $A_i$ is compared with profile limits defining classes until the condition $A_i$S$P_j$ is satisfied; (2) assigning alternative $A_i$ to class $C_{h+1}$.

According to the optimistic (or disjunctive) approach, an alternative $A_i$ is sorted into class $C_h$ if the condition $P_j$S$A_i$ is met. The procedure is implemented through two steps: (1) sequentially comparing each alternative with class boundaries, where $A_i$ is compared with profile limits defining classes until the condition $P_j$S$A_i$ is fulfilled; (2) assigning alternative $A_i$ to class $C_h$.

3.2. VIKOR Evaluating Barriers within Classes

The VIKOR methodology stands as a valuable approach for tackling the complexities inherent in MCDM problems. Its central objective is to pinpoint the most optimal compromise solution within a specified set of alternatives. This methodology proves particularly advantageous when dealing with conflicting criteria, where various alternatives may excel in certain aspects but lag behind in others. The technique places a strong emphasis on the evaluation and selection process, focusing on the proximity of alternatives to the ideal level for each criterion. By evaluating how closely alternatives approach the ideal solution, the VIKOR method provides managers with a practical and functional decision-making framework. The implementation of the VIKOR method involves the following steps.

• Formation of the input decision matrix. The input decision matrix is formulated by gathering the n alternatives, the m criteria, and the evaluations of alternatives $x_{ij}$ under each criterion.
• Standardizing the decision matrix. Normalization of the values in the decision matrix is undertaken to guarantee comparability among the criteria and establish a consistent scale, achieved through the following formula:

  $$f_{ij}\left(x\right)=\frac{x_{ij}}{\sqrt{{\sum }_{i=1}^n{x}_{ij}^2}}.$$

  (5)
• Identifying the optimal and suboptimal performance for each criterion involves determining the ideal and anti-ideal solutions, representing the best and worst performance levels, respectively, for each criterion, as elaborated below. If the criterion exhibits a positive preference direction, then:

  $$f_j^{*}=max{f}_{ij};f_j^{-}=min{f}_{ij};j=1\dots m.$$

  (6)
  If the criterion has a negative preference direction, then

  $$f_j^{*}=min{f}_{ij};f_j^{-}=max{f}_{ij};j=1\dots m.$$

  (7)
  The positive ideal solution $f^{*}$ and negative ideal solution $f^{-}$ can be expressed as follows:

  $$f^{*}=\{f_1^{*},f_2^{*},\dots ,f_m^{*}\};$$

  (8)

  $$f^{-}=\{f_1^{-},f_2^{-},\dots ,f_m^{-}\}.$$

  (9)
• Computing the collective utility and individual regret is undertaken to empower decision-makers in discerning the alternative that attains the highest overall performance (group utility, $S_i$), simultaneously excelling in individual criteria (minimal individual regret, $R_i$). These metrics are computed as follows:

  $$S_i=\sum _{j=i}^m{w}_j\frac{f_j^{*}-f_{ij}}{f_j^{*}-f_j^{-}};$$

  (10)

  $$R_i=max\left[w_j\frac{(f_j^{*}-f_{ij})}{(f_j^{*}-f_j^{-})}\right];$$

  (11)
  $w_j$ being the weight assigned to criterion j.
• Achieving the VIKOR index for each alternative involves calculating the composite measure $Q_i$, which considers both the group utility (proximity to the ideal solution) and individual regret (deviation from the best performer for each criterion). The VIKOR index is computed using the following formula:

  $$Q_i=\gamma \frac{S_i-S^{*}}{S^{-}-S^{*}}+(1-\gamma )\frac{R_i-R^{*}}{R^{-}-R^{*}};$$

  (12)

  where $\gamma =0.5$ indicates the maximum group utility, $S^{*}=min\left\{S_i\right\}$, $S^{-}=max\left\{S_i\right\}$, $R^{*}=min\left\{R_i\right\}$, and $S^{-}=max\left\{R_i\right\}$.
• Prioritizing alternatives according to their S, R, and Q values. The ranking is achieved by organizing these values in descending order, guaranteeing that the alternative with the smallest VIKOR value receives the highest rank. The result yields three separate ranking lists.
• Recommending a compromise solution. This proposed solution typically entails choosing the alternative that attains the highest rank in the overall ranking, considering all three values (S, R, and Q). This alternative signifies the most balanced compromise, encompassing considerations of group utility, individual regret, and overall performance.

4. Application of the Proposed Methodology: The Case of an Italian Automotive Paint Shop

This case study examines the Italian automotive paint shop analyzed in [3], specializing in the execution of high-quality auto-painting processes. The primary objective of the company is to execute the auto-painting work cycle, taking into account the journey from the initial acceptance phase to the delivery of the finished vehicle.

This work cycle is methodically organized into distinct steps: acceptance, preparation, cabin painting, finishing, and delivery. In the acceptance phase, the vehicle undergoes scrutiny, documentation, and acceptance for painting. Preparation involves rendering the vehicle paint-ready through cleaning, sanding, and masking to ensure an even and smooth surface. Cabin painting is the central painting phase where the actual application of paint transpires, utilizing appropriate products and techniques, which may involve spraying guns or automated systems. Finishing is the post-painting phase that concentrates on refining the painted surface through activities like polishing to achieve the desired appearance and texture. Delivery, the final phase, encompasses the preparation of the vehicle for customer handover, involving quality checks, documentation, and the ultimate transfer of the vehicle to its owner. In each phase of the work cycle, activities are linked to clearly defined tasks, each one of them requiring specialized skills, equipment, and specific materials. This often necessitates the involvement of multiple trained workers. Effective coordination and communication within this workforce are imperative for ensuring the seamless, efficient, and productive execution of tasks. The equipment used within the auto-painting facility features a painting booth, which is an enclosed monoblock chamber meticulously designed for vehicle painting. The booth is equipped with robust ventilation and aspiration systems designed to eliminate paint fumes and overspray, thereby guaranteeing a clean and secure work environment. The construction of the room involves materials capable of withstanding the elevated temperatures generated during the paint-drying process and the chemical properties of the paints and coatings used in auto-painting. Adherence to the manual of use and maintenance is essential, providing comprehensive guidelines for the proper operation and upkeep of the booth and ensuring its optimal functionality and durability. Regular maintenance and cleaning of the booth are imperative to prevent malfunctions or accidents that could compromise the quality and safety of the auto-painting process. Diligent usage and maintenance of the painting booth play a pivotal role in the efficient operation of the auto-painting facility, culminating in the delivery of high-quality results.

A comprehensive examination and prioritization of risks were conducted, as detailed in [3], revealing the significant presence of the risk associated with a polluted area. In light of this finding, the company’s management strategically formulated an approach to proactively address environmental concerns. This proactive strategy involves the implementation of an environmental monitoring program that is meticulously designed for the surrounding area of the painting booth. To facilitate the company in achieving this strategic objective, the current application adopts a systematic approach. Initially, it categorizes the barriers outlined in Table 1 based on their complexity levels, employing the ELECTRE TRI method. Following this classification, the application further refines the decision-making process by prioritizing these barriers within each designated category, utilizing the VIKOR method. This dual-method approach ensures a comprehensive and targeted strategy for overcoming barriers and attaining the company’s overarching sustainability goals.

In implementing the integrated MCDM approach, the key role of expert interviews has to be underlined. A decision-making team comprising three experts from the company was instrumental in establishing coherent criteria, laying the groundwork for the practical application of the proposed methodology. Each of the involved stakeholders contributed unique skills and perspectives, making the whole decision-making process reliable and ensuring adherence to the practical reality under study. The first expert, specializing in risk analysis and management, provided invaluable insights into potential barriers and their implications, ensuring a comprehensive understanding of the decision-making landscape. With a solid background in environmental monitoring, the second expert offered a nuanced perspective on the specific challenges related to the company’s objectives, providing insightful details about crucial environmental considerations. Lastly, the third expert, skilled in operational execution and project management, approached the process with pragmatism, ensuring that the proposed methodology aligned seamlessly with practical implementation strategies. Led by structured interviews, these experts engaged in in-depth discussions, sharing their expertise and collectively shaping coherent criteria for the integrated MCDM approach. Their collaborative efforts enhanced the robustness of the methodology, ensuring its alignment with the company’s objectives and operational realities, thereby paving the way for effective decision-making support within the Italian automotive paint shop object of the present case study. The criteria, collaboratively formulated with the input of decision-makers, are described in

Table 2. Set of criteria chosen in cooperation with the decision-making team.

ID	Criterion	Description
B1	Impact Severity	It evaluates the potential severity of each barrier’s impact on the overall effectiveness of the monitoring program while considering both short-term and long-term consequences. It also determines how disruptive each barrier could be to the implementation process and ongoing operations. This criterion quantifies the financial impact of each barrier, including initial costs, operational expenses, and potential losses.
B2	Complexity of Mitigation	It assesses the complexity with which solutions or mitigation strategies can be identified for each barrier. Some barriers may have straightforward solutions, while others may require more complex interventions. This criterion evaluates the resources (financial, technological, human) needed to implement and sustain mitigation measures for each barrier while considering the time required to address and mitigate each barrier. Some barriers may need immediate attention, while others may have a more extended time-frame for resolution.
B3	Misalignment with Objectives	This criterion evaluates how poorly the proposed mitigation aligns with the overall goals and objectives of the environmental monitoring program. It assesses whether the proposed solutions align with regulatory requirements and standards and considers the risk that the proposed solutions do not contribute to the long-term sustainability and success of the monitoring program.
B4	Risk Likelihood and Frequency	It estimates the likelihood of each barrier occurring based on historical data, industry trends, or expert opinions. This criterion considers how frequently the barrier might occur and whether it represents a one-time event or a recurring issue. It lastly identifies potential early warning signs or indicators that may help anticipate and proactively address each barrier before it becomes a significant problem.

.

For practical use of the proposed MCDM framework, a need is specified for companies to carefully select team members based on their expertise, experience, and relevance to the objectives of the decision-making process, thereby ensuring comprehensive and informed decision-making.

4.1. ELECTRE TRI Application

The first step carries out the ELECTRE TRI application. The preliminary collection and formalization of some input information is required for the practical implementation of the technique. In particular, the following is established:

• The impact severity, complexity of mitigation, misalignment with objectives, and risk likelihood and frequency are the four criteria $B_k,(k=1,\dots ,4)$, each one of them having an associated $w_k=\frac{1}{4}$ as a weight expressing same mutual importance;
• The barriers are sorted into three classes representing different levels of complexity, namely A (high complexity), B (medium complexity), and C (low complexity), each of these classes being determined by two reference profiles;
• Each criterion is characterized by the two previously mentioned reference profiles $P_j,(j=1,\dots ,2)$ corresponding to specific evaluations;
• The alternatives to be sorted by ELECTRE TRI are the thirty barriers described in Table 1, each one characterized by quantitative evaluations under each criterion, as established by the involved decision-making team (see Table 4, columns 3–6);
• The cutting value $\lambda$, that is, the threshold value defined in the last point of Section 3.1, is herein assumed to be equal to 0.8;
• The indifference ($I_k$), strong preference ($S_k$), and veto ($V_k$) thresholds, characterizing outranking relations, are defined for each criterion.

The described information, quantified according to the procedure explained in [59], is synthesized in

Table 3. ELECTRE TRI input parameters.

	B1	B2	B3	B4
P2	7.33	6.33	6.33	7.00
P1	5.67	4.67	4.67	5.00
$I_k$	0.28	0.28	0.28	0.33
$S_k$	0.42	0.42	0.42	0.50
$V_k$	0.83	0.83	0.83	1.00
$w_k$	0.25	0.25	0.25	0.25

. The barriers formalized in Table 1 have been quantitatively evaluated by the previously presented decision-making team by providing a numerical score on a scale (1–10) under each of the criteria defined in Table 2. In detail, the data collection process involved a complementary approach to gathering insights and perspectives. Expert opinions were once again elicited through structured interviews with the decision-making team, ensuring that their extensive knowledge and experience were effectively incorporated and quantitatively translated. Interviews were conducted in a collaborative and iterative manner, allowing for in-depth discussions and the exploration of diverse viewpoints. Additionally, various brainstorming sessions were organized, providing a platform for the decision-making team to collectively brainstorm ideas and refine the provided evaluations. The integration of expert opinions and brainstorming sessions ensured that the data collection process was comprehensive and holistic, capturing a wide range of insights and achieving consensus on the final evaluations. Quantitative evaluations of barriers under criteria are given in

Table 4. Barriers identified in the literature (Table 1) and their evaluation and assignment.

ID	Barrier	B1	B2	B3	B4	Pessimistic	Optimistic
FB1	Limited Budget	7	5	6	4	C	A
FB2	High Initial Investment	9	8	7	8	A	A
FB3	Operational Costs	6	4	5	3	C	C
FB4	Cost of Expertise	8	7	6	7	A	A
FB5	Unforeseen Expenses	5	6	7	4	C	A
TC1	Complex Equipment	9	8	7	9	A	A
TC2	Integration Issues	7	6	5	6	B	B
TC3	Data Management	5	4	3	4	C	C
TC4	Technological Obsolescence	8	7	6	8	A	A
TC5	Training Requirements	6	5	4	5	C	C
RC1	Changing Regulations	8	7	6	8	A	A
RC2	Legal Complexity	9	8	7	9	A	A
RC3	Documentation Requirements	5	4	4	4	C	C
RC4	Penalties for Non-Compliance	7	6	6	7	A	A
RC5	Conflict of Standards	4	3	3	3	C	C
HRC1	Expertise Shortage	9	8	7	9	A	A
HRC2	Workforce Training	6	5	4	5	C	C
HRC3	Employee Resistance	4	3	3	3	A	A
HRC4	Staff Turnover	8	7	6	8	B	B
HRC5	Competing Priorities	5	4	5	4	B	B
CR1	Data Interpretation	5	4	3	4	A	A
CR2	Timely Reporting	7	6	5	7	A	A
CR3	External Communication	8	7	6	8	A	A
CR4	Language Barriers	6	5	4	6	B	B
CR5	Stakeholder Engagement	9	8	7	9	B	B
PRRM1	Negative Public Perception	8	7	7	8	A	A
PRRM2	Reputation Risks	9	8	8	9	A	A
PRRM3	Media Scrutiny	7	6	6	7	A	A
PRRM4	Competitive Disadvantage	6	5	5	6	B	B
PRRM5	Community Relations	9	8	8	9	B	B

, along with ELECTRE TRI results in terms of assignment to complexity classes (A, B, and C) through both the pessimistic and optimistic procedures.

The ELECTRE TRI results displayed in the last two columns of Table 4 were achieved by treating the data collected in Table 3 and the quantitative evaluations reported in Table 4 and eventually double-checking via the J-Electre-v3.0 software for multi-criteria decision aid [60].

In the context of ELECTRE TRI, a preference for results derived through a pessimistic procedure is justified by its more conservative nature. The pessimistic approach indeed tends to prioritize cautious and risk-averse decision-making, making it well suited for situations where the consequences of potential errors are significant. By adopting results derived from the pessimistic procedure, decision-makers aim to assume a more cautious perspective, ensuring that the assignment of alternatives to classes is robust and less susceptible to uncertainties or variations in the input data. In the subsequent application of VIKOR, results obtained through the pessimistic procedure are given precedence, while the categorization derived via the optimistic procedure is excluded. This choice ensures that the decision-making process remains aligned with a more conservative perspective, enhancing the robustness and reliability of the final decisions in the face of real-world complexities.

4.2. VIKOR Application

Table 5. Normalized decision matrix: class A (high complexity).

	B1	B2	B3	B4
FB2	0.298	0.303	0.294	0.272
FB4	0.265	0.265	0.252	0.238
TC1	0.298	0.303	0.294	0.306
TC4	0.265	0.265	0.252	0.272
RC1	0.265	0.265	0.252	0.272
RC2	0.298	0.303	0.294	0.306
RC4	0.232	0.227	0.252	0.238
HRC1	0.298	0.303	0.294	0.306
HRC3	0.132	0.114	0.126	0.102
CR1	0.165	0.151	0.126	0.136
CR2	0.232	0.227	0.21	0.238
CR3	0.265	0.265	0.252	0.272
PRRM1	0.265	0.265	0.294	0.272
PRRM2	0.298	0.303	0.336	0.306
PRRM3	0.232	0.227	0.252	0.238

,

Table 6. Normalized decision matrix: class B (medium complexity).

	B1	B2	B3	B4
TC2	0.363	0.359	0.323	0.321
HRC4	0.415	0.419	0.387	0.428
HRC5	0.259	0.239	0.323	0.214
CR4	0.311	0.299	0.258	0.321
CR5	0.467	0.479	0.452	0.481
PRRM4	0.311	0.299	0.323	0.321
PRRM5	0.467	0.479	0.516	0.481

and

Table 7. Normalized decision matrix: class C (low complexity).

	B1	B2	B3	B4
FB1	0.445	0.386	0.452	0.348
FB3	0.381	0.309	0.377	0.261
FB5	0.318	0.463	0.528	0.348
TC3	0.318	0.309	0.226	0.348
TC5	0.381	0.386	0.302	0.435
RC3	0.318	0.309	0.302	0.348
RC5	0.254	0.231	0.226	0.261
HRC2	0.381	0.386	0.302	0.435

report the normalized decision matrices for each class via Formula (5).

The VIKOR procedure was carried out and results were eventually double-checked through the 2023 OnlineOutput MCDM software by attributing the same importance to the four criteria, as reported in

Table 8. VIKOR ranking list: class A (high complexity). The compromise solution is in bold.

	R Value	Rank in R	S Value	Rank in S	Q Value	Rank in Q
FB2	0.050	2	0.092	3	0.146	3
FB4	0.100	3	0.283	6	0.342	6
TC1	0.050	2	0.050	2	0.125	2
TC4	0.100	3	0.242	5	0.321	5
RC1	0.100	3	0.242	5	0.321	5
RC2	0.050	2	0.050	2	0.125	2
RC4	0.100	4	0.383	7	0.392	7
HRC1	0.050	2	0.050	2	0.125	2
HRC3	0.250	6	1.000	10	1.000	10
CR1	0.250	6	0.858	9	0.929	9
CR2	0.150	5	0.433	8	0.517	8
CR3	0.100	3	0.242	5	0.321	5
PRRM1	0.050	2	0.192	4	0.196	4
PRRM2	0.000	1	0.000	1	0.000	1
PRRM3	0.100	4	0.383	7	0.392	7

,

Table 9. VIKOR ranking list: class B (medium complexity). The compromise solutions are in bold.

	R Value	Rank in R	S Value	Rank in S	Q Value	Rank in Q
TC2	0.188	4	0.588	4	0.688	4
HRC4	0.125	3	0.300	3	0.410	3
HRC5	0.250	6	0.938	7	1.000	7
CR4	0.250	6	0.775	6	0.913	6
CR5	0.062	2	0.062	2	0.158	2
PRRM4	0.188	5	0.712	5	0.755	5
PRRM5	0.000	1	0.000	1	0.000	1

and

Table 10. VIKOR ranking list: class C (low complexity). The compromise solution is in bold.

	R Value	Rank in R	S Value	Rank in S	Q Value	Rank in Q
FB1	0.125	1	0.271	1	0.000	1
FB3	0.250	4	0.625	4	0.743	5
FB5	0.167	2	0.292	2	0.181	2
TC3	0.250	4	0.708	6	0.800	6
TC5	0.188	3	0.354	3	0.307	3
RC3	0.188	3	0.646	5	0.507	4
RC5	0.250	4	1.000	7	1.000	7
HRC2	0.188	3	0.354	3	0.307	3

.

These tables show the obtained S (group utility), R (individual regret), and Q (VIKOR index) values, from which we can derive the final compromise solutions, highlighted in bold in the tables, as clarified next.

The alternative $A^{\left(1\right)}$, achieving the highest rank based on the minimum Q measure, proves to be the most suitable compromise solution only if both of the following conditions are met.

• Condition 1 (acceptable advantage): $Q\left(A^{\left(2\right)}\right)-Q\left(A^{\left(1\right)}\right)\ge \frac{1}{(n-1)}$, where $A^{\left(1\right)}$ represents the alternative ranked first according to Q, $A^{\left(2\right)}$ denotes the alternative ranked second in the Q-based ranking list, and n stands for the total number of alternatives.
• Condition 2 (acceptable stability): Additionally, alternative $A^{\left(1\right)}$ must also achieve the highest rank based on S, R, or both.

If either of the conditions is not met, a set of compromise solutions is proposed, comprising the following.

• Solution 1: Alternatives $A^{\left(1\right)}$, $A^{\left(2\right)}$, …, $A^{\left(n\right)}$ are chosen if the condition of acceptable advantage (condition 1) is not fulfilled. Here, alternative $A^{\left(n\right)}$ is selected by ensuring $Q\left(A^{\left(n\right)}\right)-Q\left(A^{\left(1\right)}\right)\ge \frac{1}{(n-1)}$ for maximum n, being the positions of these alternatives defined as ”in closeness”.
• Solution 2: Alternatives $A^{\left(1\right)}$ and $A^{\left(2\right)}$ are chosen if the condition of acceptable stability (condition 2) is not met.

We can observe that we have a unique compromise solution for high-complexity class A (PRRM₂, reputation risk) and low-complexity class C (FB₁, limited budget). This is why these solutions simultaneously respect the two conditions highlighted before. In the case of medium-complexity class B, we have two resulting compromise solutions, these are PRRM₅ (community relations) and CR₅ (stakeholder engagement). This is why solution PRRM₅, which is the solution with the first rank in Q, does not respect the condition of acceptable advantage.

5. Discussion of Results and Managerial Implications

Implementing a monitoring system to mitigate the risk of a polluted area around the cabin paint booth in automotive paint shops is a complex undertaking, with barriers varying in intensity across different complexity classes. According to the pessimistic procedure of the ELECTRE TRI, fifteen barriers were allocated to the high-complexity class (class A), seven barriers to the medium-complexity class (class B), and eight barriers to the low-complexity class (class C), providing the following comprehensive breakdown.

• High-complexity class A

  -

  FB₂, High Initial Investment

  -

  FB₄, Cost of Expertise

  -

  TC₁, Complex Equipment

  -

  TC₄, Technological Obsolescence

  -

  RC₁, Changing Regulations

  -

  RC₂, Legal Complexity

  -

  RC₄, Penalties for Non-Compliance

  -

  HRC₁, Expertise Shortage

  -

  HRC₃, Employee Resistance

  -

  CR₁, Data Interpretation

  -

  CR₂, Timely Reporting

  -

  CR₃, External Communication

  -

  PRRM₁, Negative Public Perception

  -

  PRRM₂, Reputation Risks

  -

  PRRM₃, Media Scrutiny

• Medium-complexity class B

  -

  TC₂, Integration Issues

  -

  HRC₄, Staff Turnover

  -

  HRC₅, Competing Priorities

  -

  CR₄, Language Barriers

  -

  CR₅, Stakeholder Engagement

  -

  PRRM₄, Competitive Disadvantage

  -

  PRRM₅, Community Relations

• Low-complexity class C

  -

  FB₁, Limited Budget

  -

  FB₃, Operational Costs

  -

  FB₅, Unforeseen Expenses

  -

  TC₃, Data Management

  -

  TC₅, Training Requirements

  -

  RC₃, Documentation Requirements

  -

  RC₅, Conflict of Standards

  -

  HRC₂, Workforce Training

In the high-complexity class (class A), challenges such as a high initial investment, expertise costs, and evolving regulations pose significant obstacles. Addressing these issues necessitates a strategic approach, including resource allocation, workforce planning, and robust compliance strategies. Moreover, the existence of barriers such as technological obsolescence and employee resistance emphasizes the need for ongoing adaptation and proactive employee engagement initiatives. Through these concerted efforts, organizations can lay the groundwork for sustainable growth and resilience. Moving to the medium-complexity class (class B), barriers like integration issues, staff turnover, and competing priorities add a layer of difficulty. These challenges require collaborative solutions such as streamlined integration testing, effective employee retention strategies, and prioritization frameworks to navigate the dynamic landscape of competing demands. Furthermore, stakeholder engagement and communication strategies play a critical role in maintaining a cohesive approach across different organizational levels, ensuring alignment and synergy in addressing these challenges. In the low-complexity class (class C), challenges such as a limited budget, operational costs, and adherence to industry standards may seem less urgent, yet they remain fundamental to address. Prioritizing efficient resource allocation, rigorous financial planning, and implementing standardized documentation procedures are crucial in overcoming these challenges effectively. Moreover, addressing such issues as conflicts of standards and workforce training implies the necessity for a comprehensive and standardized approach to compliance and skill development.

The application of the ELECTRE TRI procedure to categorize barriers into management complexity classes offers significant advantages in terms of prioritization and decision-making. By systematically organizing barriers according to their complexity levels, this approach indeed provides a clear and structured framework for understanding the diverse array of challenges faced by organizations. The carried-out categorization enables managers to prioritize their focus and resources effectively, addressing high-complexity barriers that pose the most immediate and critical threats first. Additionally, by delineating barriers into distinct classes, the ELECTRE TRI procedure facilitates a nuanced understanding of the underlying factors contributing to each challenge. This deeper insight empowers management teams to tailor their strategies and interventions accordingly, devising targeted approaches that are better suited to address the specific nature of each barrier.

In this last regard, as the complexity of implementing a monitoring system in automotive paint shops is multifaceted, tailored strategies can be effectively formulated for each class. The VIKOR application revealed this, as, for each of the analyzed classes, certain barriers emerged as requiring priority attention. At a practical level, addressing these specific barriers is important to effectively manage the risk of polluted areas and enhance the overall sustainability of the automotive supply chain. Subsequently, detailed strategies targeting these identified barriers can be designed while aligning with broader sustainability objectives.

Considering the results that emerged from the application of the VIKOR technique (see Table 8, Table 9 and Table 10), in the high-complexity category, characterized by reputation risk, a focus on transparent communication and crisis preparedness emerged as useful. This would require establishing reliable communication channels both internally and externally, promoting transparency, and building trust with stakeholders. Additionally, developing robust crisis response plans is essential to ensure swift and effective action in the event of environmental incidents or other emergencies. By proactively addressing reputation risks through transparent communication and comprehensive crisis preparedness, automotive paint shops can maintain stakeholder trust and mitigate potential damage to their brand image and credibility. Moving to the medium-complexity class, which highlights challenges related to community relations and stakeholder engagement, the company may consider the implementation of proactive outreach initiatives while involving stakeholders in decision-making processes. By building positive relationships through social responsibility efforts and engaging with local communities and key stakeholders, the company can gain valuable insights, build trust, and foster collaborative relationships. Through these concerted efforts, organizations can enhance their reputation, strengthen community ties, and create shared value for all stakeholders involved. In the low-complexity class, marked by a limited budget, strategic measures include investing in cost-effective technologies, adopting lean operational practices, and cultivating collaborative partnerships within the automotive supply chain. This pragmatic approach aims to address financial constraints while ensuring the seamless integration of sustainable practices throughout the supply chain. By leveraging cost-effective technologies and streamlining operational processes, organizations can optimize resource utilization and enhance operational efficiency. Additionally, fostering collaborative partnerships within the supply chain facilitates knowledge sharing and resource pooling, enabling collective efforts towards sustainability goals.

These proposed strategies are designed to effectively implement monitoring systems, mitigate environmental risks, and enhance sustainability within automotive paint shops and the broader supply chain ecosystem, so as to cultivate a culture of sustainable practices that resonates throughout the automotive industry.

6. Conclusions

This paper addresses the pressing issue of integrating sustainable logistics into automotive manufacturing. An MCDM framework is proposed to recommend green supply chain strategies, particularly in the critical context of automotive paint shops. Specifically, by employing a dual-step MCDM approach (ELECTRE TRI and VIKOR), the paper analyzes primary barriers to implementing pollution risk-monitoring programs, categorizing and prioritizing obstacles to formulate effective green strategies for the sector. The proposed integrated MCDM framework identifies and prioritizes barriers to digital transformation, specifically in managing and monitoring pollution risks in painting booths through intelligent systems. The ultimate goal consists of recommending sustainable logistic strategies to respond to the most significant barriers to implementing monitoring systems and providing managerial insights for practitioners involved in digitizing paint shop operations.

This research aims to contribute to enhancing the resilience and long-term viability of automotive supply chains at a practical level. The findings of this research have explicit implications for the automotive paint shop setting. Implementing a monitoring system to mitigate pollution risks around the cabin paint booth in automotive paint shops represents a critical endeavor, particularly given its potential impact on environmental sustainability and regulatory compliance. This task involves navigating through the previously mentioned barriers, which have been herein categorized into high-, medium-, and low-complexity classes (A, B, and C). For instance, class A faces challenges such as high initial investment, expertise costs, and changing regulations, demanding strategic approaches. These challenges directly affect the feasibility and implementation of monitoring systems within the paint shop setting. Addressing these barriers effectively requires the development of tailored strategies focused on optimizing resource allocation and leveraging expertise in environmental monitoring. Class B encounters barriers like integration issues, staff turnover, and competing priorities, requiring collaborative solutions and stakeholder engagement. In the automotive paint shop context, overcoming these barriers is essential for ensuring the seamless integration of monitoring systems into existing operational processes. This involves fostering a culture of collaboration and communication among stakeholders to address common challenges and streamline implementation efforts. Class C, characterized by limited budgets, operational costs, and adherence to standards, demands efficient resource allocation and standardized procedures. In the automotive paint shop setting, where financial resources may be constrained, strategic investments in cost-effective technologies and lean practices become imperative. Moreover, ensuring compliance with regulatory standards requires the establishment of standardized procedures and continuous monitoring to mitigate potential risks.

By implementing tailored strategies, such as crisis preparedness, transparent communication, proactive outreach, and strategic investments in technology and partnerships, automotive paint shops can effectively address the identified barriers and enhance their sustainability practices. Crisis preparedness ensures that paint shops are equipped to respond swiftly and effectively to any unforeseen environmental incidents, minimizing potential damages and maintaining regulatory compliance. Transparent communication is vital in building trust and credibility both internally among staff and externally with stakeholders and regulatory bodies. By fostering open and transparent communication channels, paint shops can promote accountability and collaboration, facilitating the smoother implementation of monitoring systems and garnering support for sustainability initiatives. Proactive outreach efforts involve actively engaging with stakeholders, including employees, local communities, and regulatory agencies, to solicit feedback, address concerns, and foster a sense of shared responsibility towards environmental stewardship. This proactive approach not only enhances the effectiveness of monitoring systems but also strengthens relationships and fosters a positive reputation for the paint shop within the community. Furthermore, strategic investments in technology and partnerships enable paint shops to leverage innovative solutions and expertise to overcome barriers and optimize their environmental performance. Whether through the adoption of advanced monitoring technologies or forging strategic partnerships with environmental consultants or technology providers, these investments can yield significant long-term benefits in terms of efficiency, effectiveness, and regulatory compliance.

The practical implications of this research offer actionable insights that can drive meaningful change and foster sustainability in automotive paint shop operations.