Next Article in Journal
A Novel Model Structured on Predictive Churn Methods in a Banking Organization
Next Article in Special Issue
Attributes of Business Incubators: A Conjoint Analysis of Venture Capitalist’s Decision Making
Previous Article in Journal
Does Fixed Income Buffer against Fraud Shocks?
Previous Article in Special Issue
Auditor Judgements after Withdrawal of the Materiality Accounting Standard in Australia
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Framework on Performance Management in Automotive Industry: A Case Study

by
Elena Lascu
1,
Irina Severin
1,*,
Florina Daniela Lascu
1,
Razvan Adrian Gudana
1,
Gabriela Nalbitoru
1 and
Nicoleta Daniela Ignat
2
1
Faculty of Industrial Engineering & Robotics, University POLITEHNICA Bucharest, 060042 Bucharest, Romania
2
Faculty of Entrepreneurship, Business Engineering and Management, University POLITEHNICA Bucharest, 060042 Bucharest, Romania
*
Author to whom correspondence should be addressed.
J. Risk Financial Manag. 2021, 14(10), 480; https://doi.org/10.3390/jrfm14100480
Submission received: 18 August 2021 / Revised: 8 October 2021 / Accepted: 11 October 2021 / Published: 12 October 2021
(This article belongs to the Special Issue Advances in International Management Research)

Abstract

:
The purpose of this research is to identify the risks and deficiencies that affect the performance of companies that provide vehicle after-sales services. Thus, this paper highlights the results of a comparative study based on a questionnaire conducted at the level of six brands in the automotive industry. A model was developed to investigate the factors that affect the global performance of the after-sales sector and the authenticity of the information related to the issue studied. Moreover, based on the collected data, this study evaluates the strategies related to performance management used by the organizations studied. In the end, even if the results showed a score of 81% on the questionnaire, we found that companies that provide vehicle after-sales services have not implemented and do not maintain totally the strategies related to performance management. Consequently, the need for change can be emphasized. Based on the analyzed data in the second part of the paper, we identified deficiencies and risks in terms of the organization, operation and management of the service units. These results confirm that the vehicle repair service has a significant influence on employee and customer satisfaction, on the quality of the vehicles repaired and the repair completion time.

1. Introduction

This long transition period, without a consistent strategy and plan through which the world economy has passed (Staš et al. 2015), has affected all areas of activity, and the vehicle industry is no exception. According to statistical research (Tigist and Nicolescu 2018), the number of cars has increased from year to year and the need for vehicle maintenance and repair services has increased in line with this phenomenon.
Currently, in order to survive on the market, it is necessary for these companies to improve their competitiveness (Johannsen 2013; Lee and Seo 2017) in order to perform and even to achieve excellence (Molina-Azorín et al. 2009). Improving the quality of repair services, managing the process for transformation and adaptation, as well as rigorous management, are the key elements in meeting customer requirements. According to Chelcea S., for the automotive services sector, these requirements mean quality orientation and involve Total Quality Management tools and methods (Chelcea 2001). Similar ideas have been presented by the authors Adèr et al. (2008), Chardine-Baumann and Botta-Genoulaz (2014), Fechete (2016) and Nenadál et al. (2018).
Despite the appearance of more and more small- and medium-sized companies for vehicle after-sale services, many previous studies have focused on large companies. In consequence, this study proposes a framework to help companies that provide vehicle after-sale services to grow in order to survive in this competitive environment (Jeong et al. 2016; Najwa et al. 2019; Roth and Manor 2009; Werner-Lewandowska and Golinska-Dawson 2021; Wismantoro et al. 2019; Young et al. 2017).
For several years now, in the vehicle after-sale sector, a decrease in the overall performance has been observed at the national level. The problems and challenges faced by these companies indirectly and implicitly affect customers and employees. These companies are aware that quality repair services led customers to loyalty and redemption, but they nevertheless cannot achieve success if they do not take into account the needs of all stakeholders in the design and delivery of services (Ling et al. 2016).
In these conditions, in order to determine the elements that lead to the unsatisfactory overall performance of companies within the vehicle after-sale services sector, the purpose of this research is to assess the current state of the implementation of the strategies related to performance management and to identify the main challenges that these organizations are facing. Process management aims to monitor and control the processes defined in an organization (ISO 9001:2015 2005). Quality management practices, effectively supported by top management, lead to process efficiency (Nguyen et al. 2018). Performance management may include essential elements and management practices to achieve the set objectives (Molina-Azorín et al. 2009; To et al. 2018). Understanding and managing the correlated processes within a management system supports an increase in an organization’s efficiency, effectiveness and resources economy, the final aim being to obtain the desired results (ISO 9001:2015 2005).
Determining the elements that lead to unsatisfactory overall performance, as well as the main challenges faced by these vehicle after-sale services organizations, is a fundamental strategy, bringing important benefits to the organizations studied in terms of time optimization, cost reduction (Malevskaia-Malevich et al. 2018) and improvements in general activity (Medne and Lapina 2019).
The current study carries out the performance evaluation of vehicle after-sale services using data from the six main car brand companies in Romania. More precisely, for each car brand company presented in this study, we consider all their after-sales service divisions in the Bucharest area.
We use a comparative methodology in the research, because this approach is an effective tool for analysis that serves as a starting point for the detailed examination of the similarities and differences between the comparative assessment group and the specific group (Proksch et al. 2019).
In the first step, the literature and the market are reviewed and a model of the strategies related to performance management is developed. In the second step, after analyzing the products, services and business, the organizations are selected and grouped based on their competence and the conditions in the foreign market. In the third step, based on the types of organizations previously identified, the working hypotheses are set and the questionnaire is prepared and distributed. In the fourth step, the comparative analysis is performed based on the data collected. The fifth step represents discussions of the results and conclusions on the research. The results of the research will serve to formulate recommendations and strategic directions for improving vehicle repair services.
This research has not only theoretical significance but also positive practical significance that will provide vehicle repair services organizations with a good framework to identify the causes underlying business dysfunctions.
The remainder of the paper is organized as follows: presents the literature review, describes the research methodology, presents the data analysis, explains the results and provide comments on the results and conclusions.

2. Literature Review

In recent years, the business performance and sustainability interdisciplinary research has been increasingly expanded (Mitchell and Jolley 2001). However, the automotive industry and the vehicle after-sale services sector has not been systematically explored yet.
In today’s economy, the business environment is becoming increasingly aware of the need for sustainability management that aims at the simultaneous achievement of social, environmental and economic performance (EFQM 2020). Performance evaluation is essential for any organization in the automotive industry in managing operations as it provides a way to improve operations for business sustainability.
Sustainability performance is seen as a result of sustainability management. Sustainability performance is defined as “the performance of a company in all dimensions and for all motor factors” (Schaltegger and Wagner 2006). In 2014, Chardine-Baumann E. defined sustainability performance as “the combination of economic, social and environmental performance” (Chardine-Baumann and Botta-Genoulaz 2014). Isaksson (2006) and Siva et al. (2016) presented similar ideas.
In recent decades, the methods used for collecting information regarding key performance indicators have included performance reviews, interviews, questionnaires and surveys or benchmarking (ISO/TR 10014:2021 2021; Hong and Kim 2020; Hong et al. 2020; ISO 9004:2018 2005; Khoshkangini et al. 2020). However, studies related to the performance of vehicle after-sale services are very limited in past research.
The outputs of analysis can be used for internal (between activities and processes) or external (between other organizations) benchmarking to show the trends over time (ISO 9004:2018 2005).
Starting from existing research in similar fields (Akhmetshin et al. 2019; Draghici et al. 2014; Liu et al. 2017; Yee et al. 2013) and quality management guidelines (ISO 10006:2005 2017; ISO/TR 10014:2021 2021), this study focuses on internal management, with an emphasis on process and service management, strategies, human resources, customer management and the results obtained.
Popescu et al. (2015) stated that “initiative, creativity, innovation, development of organizational capacity and risk-taking capacity are also attributes of progress capable of generating national wealth”. The authors Parameshwaran and Srinivasan (2008), Olaru et al. (2014) and Nenadál et al. (2018) presented similar ideas.
Our research aimed to determine the factors that influence the performance of organizations (Okolnishkova et al. 2018) and to evaluate the strategies related to performance management used by these organizations by proposing a framework. The study was conducted on a sample of six brands in the automotive industry. The analyzed industry is important both economically and socially and plays an important role in the daily life of the company, employees and customers.

3. Research Methodology

This paper is part of a larger study and is based on the current situation of the vehicle after-sale services sector in the automotive industry, which has reported that its overall performance is below target. These results are reflected in the low level of financial performance and customer and employee satisfaction.
In order to achieve the proposed goal, i.e., the identification of critical factors that affect business performance, several steps were completed (Jiang and Prabhu 2008; Mitreva et al. 2019). The specific steps were as follows: establishing the general topic, defining the research problem, setting the research objective, elaborating the question to be answered by the study, formulating the research hypotheses and testing the hypotheses, as shown in Figure 1.
Documentary research: The analysis of social documents is a method of observation that allows the interrogation of a variety of sources of information. The studied social documents were written (official records) and unwritten (news, information, newspapers, magazines).
Survey: The social survey is a statistical method and involves the collection of information related to stakeholders in order to identify relationships and the statistical distribution of elements of interest. The tool used in this method is the questionnaire.
Given the current stage of the research, the basic tool used was the questionnaire. The reason behind the decision to apply the questionnaires was based on the idea that the information obtained is quantitative and qualitative and that stakeholders can express themselves directly. Moreover, by applying the questionnaires, it was possible to deepen aspects that could not be accessed through other sources.
From a sociological point of view, Chelcea (2004) offers a broader definition of the questionnaire as “a technique and, accordingly, an investigative tool consisting of a set of written questions and possibly graphic, orderly and psychological images, which, by being administered by the survey operators or by self-administration, determine the following answers from the respondents to be recorded in writing”.
The questionnaire has advantages over the interview because it is cheap, contains standardized answers and requires little effort from the subjects. According to Catoiu (2002), the questionnaire represents “a formalized set of questions, designed to generate the data needed to achieve the objectives of a marketing research”.
The objectives of the questionnaires applied to the vehicle repair service organizations were as follows:
-
Identifying basic information related to the studied subject;
-
Highlighting the relationship between performance and other variables;
-
Verification of the authenticity of the information collected from the studied social documents: financial performance, results, management and organization, other problems faced by the organization.

3.1. Methodology

To achieve the research objectives, this study proposes a framework based on a questionnaire (Munn and Drever 2004), to help small and medium-sized companies with vehicle after-sale services to grow in order to survive in the competitive environment. The primary analysis of the applied questionnaires was characterized by the following elements:
-
Information related to the studied problem—identification of the factors that influence the performance of organizations in the automotive sector;
-
Sampling framework—at the headquarters of the car service organizations in Bucharest;
-
Target group—managers;
-
Sample—selection by statistical method based on reasoning—10 organizations, 10 managers;
-
Time required for designing and testing the questionnaires—January–March 2018.
The questionnaire applied in the organization was prepared on the principle of sectioning and included two sections. The first section contained 36 questions subdivided into 7 sections (S1 to S7), which were further subdivided into 5 criteria (Q1 to Q5) covering the entire field of activity. This first part of the questionnaire contained questions regarding the strategies related to performance management. The second section contained an open-ended question, to which the respondent was allowed to formulate his answer, and focused on the declared problems of the organization.
The second stage in the design of the questionnaire is summarized in Figure 2.
The questionnaire was structured according to strategies related to performance management, and the links between the analyzed criteria are presented in Figure 3.
Performance management strategies were as follows: Leadership (S1), Strategies (S2), Management Systems (S3), Customer Management (S4), Human Resources Management (S5), Risk Management (S6) and Results (S7).
Information on the evaluation rules for each section/criterion of questionnaire, as well as related to the awarding of the final score, can be found in Table 1, Table 2 and Table 3.
Based on the study of these factors specific to each brand, the quality policy statements, as well as the data presented above, we could conclude that these organizations had certain general features in common:
  • Companies were constantly expanding;
  • They had a local presence and were some of the top brands in the automotive industry;
  • They had policies regarding the implementation and maintenance of the integrated quality–security–environment management system;
  • They had a vision, mission and objectives related to the performance, which were well-defined and oriented towards quality, security and the environment; continuous improvement; customers; employees; performance; risk; safety; social responsibility.

3.2. Hypotheses

The established hypotheses represent the core of research and are a powerful tool that guides the entire research. Summing up some definitions given by other authors, the hypothesis represents “a verbal sentence, derived from observations, theories or the results of other research on the relationship between variables” (Schwab 2005), “which is expressed in the form of a prediction regarding the relationship between the variables investigated” (Spector 2012; Wagner 2021) and which “can be tested empirically” (Liu et al. 2017).
Based on the statements of the cited authors, we can conclude that the hypotheses have the role of presenting concretely what we expect to be achieved in our research.
To validate the structural research model proposed in Figure 3, as well as to provide answers to the objectives of this research, the following research hypotheses were established:
Hypothesis 1 (H1).
For the studied organizations, in full expansion, only the implementation, certification and maintenance of management systems is sufficient.
This hypothesis is related to the first section of the questionnaire, and this contained closed-ended questions, which required information on management and organization.
Hypothesis 2 (H2).
By implementing the questionnaire, we will identify deficiencies and risks.
This hypothesis is related to the second section of the questionnaire, which contained an open-ended question: the respondent was asked to describe three issues faced in the service unit. Problems were presented in the order of processing the questionnaires, and, in the case of similar types of problems, a more comprehensive reformulation of the entire set of problems was chosen.

4. Data Analysis

As the present study aimed to determine the factors that influence the performance of vehicle after-sale service organizations, the questionnaires were distributed to selected organizations.
The response rate for the three questionnaires was calculated using the formula:
R = Number   of   answers Number   of   questionnaires   sent   ×   100
As can be seen in Table 4, managers answered in a proportion of 70%. If we look at the total number of potential respondents and the response rate, it can be concluded that the interest expressed by the respondents was high, which shows that the research method of applying face-to-face questionnaires was the most appropriate. This analysis is interpreted graphically in Figure 4.
The six main car brands that completed the questionnaire are listed in Table 5. Both competitors and performance management strategies declared by them are listed in Table 6.
Table 7 presents the analysis of the answers received following the application of the questionnaires. The answers collected regarding the questions related to this hypothesis are lower.
The comparison between the strategies related to performance management and the fulfillment percentages for each strategy are presented in Table 8.
The answers collected regarding the questions related to this hypothesis are centralized in Table 9.

5. Results

The results obtained in this research were previously centralized. This section contains the graphical representation of the results and their interpretation. Figure 5 shows the global scores obtained by each enterprise.
From Figure 5a, it can be seen that the lowest score was obtained by Competitor 1 and the highest score was obtained by Competitor 5. The fulfillment percentages for the questionnaire related to each competitor were as follows: Competitor 1—64.1%, Competitor 2—84.1%, Competitor 3—78.9%, Competitor 4—82.4%, Competitor 5—95.4% and Competitor 6—81.4%. The global percentage of achievement was 81%.
From Figure 5b, it can be seen that the lowest score was obtained for the seventh criterion and the highest score was obtained for the third criterion. The fulfillment percentages for each criterion were as follows: Section 1 (Management)—86.7%, Section 2 (Strategies)—81.3%, Section 3 (Management systems)—80%, Section 4 (Customer management)—89.3%, Section 5 (Employee management)—71.3%, Section 6 (Risk management)—89.3% and Section 7 (Performances and results obtained)—69.3%.
Starting from Figure 5, an in-depth analysis was performed for each competitor in terms of the sections and criteria. Figure 6 shows the total scores obtained on each section by the enterprises. Comparing the results with the maximum scores that could be obtained (presented in Table 3), we can see, for each competitor, in which section they obtained the lowest and the highest score.
The lowest scores (weaknesses) obtained were as follows: Competitor 1—S3 (Management Systems), Competitor 2—S5 (Employee management), Competitor 3—S7 (Performances and results obtained), Competitor 4—S2 (Strategies) and S7 (Performances and results obtained), Competitor 5—S7 (Performances and results obtained), Competitor 6—S7 (Performances and results obtained).
The highest scores (strengths) obtained were as follows: Competitor 1—S6 (Risk management), Competitor 2—S6 (Risk management), Competitor 3—S4 (Customer management), Competitor 4—S1 (Management), Competitor 5—S1 (Management), S2 (Strategies), S3 (Management Systems) and S4 (Customer management), Competitor 6—S6 (Risk management).
Figure 7 shows the scores obtained by all competitors. Figure 7a shows the scores obtained by Competitor 1 in each section. We can see that the lowest score was obtained on S3 (Management Systems) and the highest score was obtained on S6 (Risk management). The fulfillment percentages for the questionnaire related to Competitor 1 were as follows: Section 1 (Management)—76%, Section 2 (Strategies)—72%, Section 3 (Management systems)—28.6%, Section 4 (Customer management)—84%, Section 5 (Employee management)—44%, Section 6 (Risk management)—92% and Section 7 (Performances and results obtained)—52%.
Figure 7b shows the scores obtained by Competitor 2 in each section. We can see that the lowest score was obtained on S5 (Employee management) and the highest score was obtained on S6 (Risk management). The fulfillment percentages for the questionnaire related to Competitor 2 were as follows: Section 1 (Management)—80%, Section 2 (Strategies)—84%, Section 3 (Management systems)—88.6%, Section 4 (Customer management)—84%, Section 5 (Employee management)—76%, Section 6 (Risk management)—92% and Section 7 (Performances and results obtained)—84%.
Figure 7c shows the scores obtained by Competitor 3 in each section. We can see that the lowest score was obtained on S7 (Performances and results obtained) and the highest score was obtained on S4 (Customer management). The fulfillment percentages for the questionnaire related to Competitor 3 were as follows: Section 1 (Management)—84%, Section 2 (Strategies)—80%, Section 3 (Management systems)—88.6%, Section 4 (Customer management)—100%, Section 5 (Employee management)—56%, Section 6 (Risk management)—92% and Section 7 (Performances and results obtained)—52%.
Figure 7d shows the scores obtained by Competitor 4 on each section. We can see that the lowest scores were obtained on S2 (Strategies) and S7 (Performances and results obtained) and the highest score was obtained on S1 (Management). The fulfillment percentages for the questionnaire related to Competitor 4 were as follows: Section 1 (Management)—100%, Section 2 (Strategies)—72%, Section 3 (Management systems)—88.6%, Section 4 (Customer management)—84%, Section 5 (Employee management)—84%, Section 6 (Risk management)—76% and Section 7 (Performances and results obtained)—72%.
Figure 7e shows the scores obtained by Competitor 5 in each section. We can see that the lowest score was obtained on S7 (Performances and results obtained) and the highest score was obtained on S1 (Management). The fulfillment percentages for the questionnaire related to Competitor 5 were as follows: Section 1 (Management)—100%, Section 2 (Strategies)—100%, Section 3 (Management systems)—100%, Section 4 (Customer management)—100%, Section 5 (Employee management)—92%, Section 6 (Risk management)—92% and Section 7 (Performances and results obtained)—84%.
Figure 7f shows the scores obtained by Competitor 6 in each section. We can see that the lowest score was obtained on S7 (Performances and results obtained) and the highest score was obtained on S6 (Risk management). The fulfillment percentages for the questionnaire related to Competitor 6 were as follows: Section 1 (Management)—80%, Section 2 (Strategies)—80%, Section 3 (Management systems)—85.7%, Section 4 (Customer management)—84%, Section 5 (Employee management)—76%, Section 6 (Risk management)—92% and Section 7 (Performances and results obtained)—72%.
Based on the data summarized in Table 9, the frequency of the main categories of problems faced by the enterprises could be represented. In Figure 8, they are grouped according to the frequency of occurrence.

6. Comments on the Results: Hypothesis Testing

6.1. Verification of Hypothesis H1

Based on the results reflected above, we find that Hypothesis H1 is not fully verified. More precisely, it is not enough to only implement one or more management systems to achieve excellence.
Additionally, if we consider the extent to which the criteria and sub-criteria of the questionnaire are met, an imbalance is found and there is a decrease in the percentages of fulfillment for several criteria, which will be analyzed in the following. For the quick and correct identification of the questions, they were coded. The coding of the questions contained information related to the question number and the questionnaire to which it belonged, being designed as follows:
*QnMQ—the coding of the questions contains information related to the number of the question and the questionnaire name, where “n” represents the number of the question in the questionnaire addressed to managers (MQ)
The management criteria results are presented in Figure 9. The percentage of fulfillment was 76.7%. The answers to the Q2MQ and Q3MQ questions are highlighted, where the management supports and motivates its employees and promotes an open dialogue at a percentage of 66.7%, while the answers to the Q4MQ* question show that only 50% of the organizations are known and appreciated for social involvement.
The strategy criteria results are shown in Figure 10. The fulfillment percentage was 73.3%. The answers to questions Q8MQ and Q9MQ were noted, where a percentage of 33% of managers stated that the policy and strategy were not reviewed periodically and the strategy was not communicated to the organization’s departments. Less than 25% stated that they partially implemented the Q8MQ and Q9MQ criteria.
The category of management systems criteria results is presented in Figure 11. The fulfillment percentage of the category was 55.6%. The answers to questions Q14MQ and Q15MQ were noted, where 16.7% of managers stated that there was no integrated approach to the improvement of the mechanisms in place, nor did they provide assurances that the integrity of the SMI would be maintained in the event of change. A percentage higher than 27.8% stated that they partially implemented the Q13MQ, Q14MQ and Q15MQ criteria.
The customer management criteria results are presented in Figure 12. The fulfillment percentage was 76.7%. The answers to the Q16MQ question were noted, where 3.3% of managers claimed that the organization did not develop and maintain long-term partnerships with customers, while a further 20% of managers stated that they partially implemented the Q19MQ and Q20MQ criteria.
The employee management criteria results are presented in Figure 13. A very important aspect in the analysis of employees’ responses shows a percentage of fulfillment of the category of only 43.3%. Most of the managers, namely 46.7%, stated that they partially implemented the following criteria: setting and evaluating the individual objectives for each employee (Q22MQ); identifying, analyzing and establishing the needs for improving the professional training of employees in order to perform their duties (Q23MQ); carrying out professional development programs (Q24MQ) and establishing indicators for measuring customer satisfaction (Q25MQ).
The risk management criteria results are presented in Figure 14. The fulfillment percentage of the category among the studied organizations was 87.5%. A further 12.5% of managers claimed that the relevance of the indicators associated with the objectives was partially re-evaluated. A very important aspect in the analysis of the answers was that, in only 16.7% of the organizations, the specific objectives were re-evaluated periodically.
The performance and results obtained are shown in Figure 15. The fulfillment percentage was 53.3%. Less than 26.6% stated that they partially implemented the following criteria: integration and acceptance of ethnic minorities and disadvantaged people (Q32MQ); development of risk/accident prevention programs for the health of customers/employees (Q33MQ); improving the reputation and image among citizens (Q34MQ) and obtaining results correlated with the set objectives (Q35MQ). A very important aspect for this category was the fact that 20% of organizations did not implement the criteria included in this category.

6.2. Verification of Hypothesis H2

In order to substantiate the answer to this problem, we analyze the results presented in Table 9.
From the analysis sorted by problem categories, as shown in Figure 8, it can be seen that a high frequency of occurrence refers to high staff turnover based on their dissatisfaction and negative feedback from customers. With a slightly lower frequency of occurrence, we found insufficient employee involvement, a lack of indicators for employee satisfaction and a lack of synchronization of expenses with receipts. These types of problems were covered by the questionnaire.
Lastly, with a lower percentage of generality, we found a lack of integrated information systems, limited tools to motivate employees, long delays in solving service interventions, a lack of a program for the integration of ethnic minorities and a lack of a program to attract young high school/college graduates. These types of problems were not covered by the questionnaire.

7. Conclusions

The main contribution of this study is the elaboration of the framework for evaluating the strategies related to performance management and identifying the risks and deficiencies that affect the performance of companies that provide vehicle maintenance and repair services.
To determine the factors that influence the performance of vehicle after-sale service organizations, the first step was to identify the research objectives and set working hypotheses. This step was followed by the establishment of the research methods and tools. The research of the subject was carried out through direct observation, as well as documentary research. The questionnaire used as a research tool for data collection was created respecting the guidelines in the literature and allowed the monitoring and direction of the collection process, as well as determining the current level of major concerns related to the development of automotive services. The research conducted allowed us to determine the current level of the major concerns regarding the development of the automotive service activity. Following the examination of the answers received from the respondents, the need for change was outlined.
On one hand, the 81% score of achievement for the questionnaire shows that the vehicle repair services units implement and maintain a quality management system. In order to remain in a stable situation, vehicle repair service units should monitor both internal indicators and indicators in terms of direct competition and should be inspired by competition to solve problems. It is not enough for organizations to implement and certify one, two or more management systems, but it is mandatory that the principles of total quality be assimilated and applied in the organizational culture, an approach on which sustainable excellence is based.
On the other hand, based on the analyzed data in the second part of the questionnaire, it could be identified that there are deficiencies and risks in terms of the organization, operation and management of service units.
The analysis presented in Figure 8 shows the main categories of problems faced by organizations: high staff turnover based on their dissatisfaction and negative feedback from customers. With a slightly lower frequency, we identified the insufficient involvement of employees, the lack of indicators of employee satisfaction and the lack of synchronization of expenses with receipts.
The application of the questionnaire as a research tool gave managers the opportunity to identify areas in imbalance and where to intervene through changes to increase the quality and performance of their car services.
Based on this analysis, we may conclude that this hypothesis is verified. More precisely, based on the evaluation of the current state of the six car brands, certain deficiencies and risks can be identified regarding the activity of the vehicle services.
Through the questionnaire, two hypotheses were tested, and the results obtained showed that, although there was a high level of fulfillment of the questionnaire, the service units must progress in order to fully assimilate the principles of total quality management. Moreover, based on the analyzed results, it was found that, in order to achieve excellence in vehicle services, it is not enough to only implement two or more management systems.
Finally, our research can provide a guide for managing organizations operating in the vehicle maintenance and repair service sector and can provide a basis for developing a model of the factors that influence the performance of these organizations. The results of this model can be used in order to improve the processes and management and implicitly to increase stakeholder satisfaction. Moreover, another possible future study could provide specified practical guidelines to facilitate the implementation of business performance and sustainability practices to support industry organizations in establishing strategies effectively.
Despite the contributions of this study, there are some limitations that suggest a path for future research. A limitation of the research may be the small sample of companies, with insufficient data to guarantee reliable results. In our future research, we will address these shortcomings. Additional data should be collected to assess the practices used in a more reliable way. As more data are collected, it will be possible to extend the approach to other industries.
Based on the results obtained in this research, we can develop cost recommendations that could be useful for the vehicle maintenance and repair service sector:
-
For the evaluation of the external condition, the organization should apply at least two evaluation methods and compare the results with those of the competitors;
-
Comparison with competitors should be a comparison based on several criteria (not only one), with the assignment of a reference value for each indicator;
-
The indicators should be selected by involving the managers at the level of each department.
The practical approach of collecting data from managers appeared difficult to conduct, but the results revealed, at a certain level, a degree of insufficient rigor in current performance evaluation procedures in service and repair units, a sector with a specific culture in the framework of the automotive industry, dominated by the national context.

Author Contributions

Conceptualization, E.L., I.S., F.D.L. and R.A.G.; methodology, E.L.; validation, I.S.; formal analysis, E.L.; investigation, E.L.; resources, E.L., F.D.L., R.A.G. and G.N.; writing—original draft preparation, E.L.; writing—review and editing, E.L.; visualization, F.D.L., R.A.G. and N.D.I.; supervision, I.S. and N.D.I. All authors have read and agreed to the published version of the manuscript.

Funding

Financial support from the Competitiveness Operational Program 2014–2020, Action 1.1.3: Creating synergies with RDI actions of the EU’s HORIZON 2020 framework program and other international RDI programs, MySMIS Code 108792, Acronym project “UPB4H”, financed by contract: 250/11.05.2020, is gratefully acknowledged.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Adèr, Herman J., Gideon J. Mellenbergh, and David J. Hand. 2008. Advising On Research Methods: A Consultant’s Companion. Huizen: Johannes van Kessel Publishing. [Google Scholar]
  2. Akhmetshin, Elvir Munirovich, Vladimir Lvovich Vasilev, Natalia Ivanovna Vlasova, Aleksandr Vasilevich Kazakov, Xenia Yurievna Kotova, and Ruslan Hizrailevich Ilyasov. 2019. Improving Management Functions at an Enterprise: Levels of the Internal Control System. Quality Access to Success 20: 39–43. [Google Scholar]
  3. Catoiu, Iacob (coord). 2002. Marketing Researches. Bucharest: Uranus Publishing House, p. 311. [Google Scholar]
  4. Chardine-Baumann, Emilie, and Valerie Botta-Genoulaz. 2014. A framework for sustainable performance assessment of supply chain management practices. Computers & Industrial Engineering 76: 138–47. [Google Scholar]
  5. Chelcea, Septimiu. 2001. Methodology of Sociological Research: Quantitative and Qualitative Methods. Bucharest: Economic Publishing House. [Google Scholar]
  6. Chelcea, Septimiu. 2004. Initiation in Sociological Research. Bucharest: Communication Publishing House, p. 105. [Google Scholar]
  7. Draghici, Anca, Anca-Diana Popescu, and Luminita Maria Gogan. 2014. A proposed model for monitoring organizational performance. Procedia—Social and Behavioral Sciences 124: 544–51. [Google Scholar] [CrossRef] [Green Version]
  8. EFQM. 2020. Available online: http://www.efqm.org (accessed on 10 February 2021).
  9. Fechete, Ec F. 2016. Research on Industrial Systems Performance Management. Doctoral thesis, Transilvania University, Brasov, Romania. [Google Scholar]
  10. Hong, Jinpyo, and Boyoung Kim. 2020. Service Quality, Relationship Benefit and Experience Value in the Auto Repair Services Sector. Journal of Open Innovation: Technology, Market, and Complexity 6: 30. [Google Scholar] [CrossRef]
  11. Hong, Jinpyo, Boyoung Kim, and Sungho Oh. 2020. The Relationship Benefits of Auto Maintenance and Repair Service: A Case Study of Korea. Behavioral Sciences 10: 115. [Google Scholar] [CrossRef] [PubMed]
  12. Isaksson, Raine. 2006. Total Quality Management for sustainable development: Process based system models. Business Process Management Journal 12: 632–45. [Google Scholar] [CrossRef]
  13. ISO 10006:2005. 2017. Quality Management Systems. Guidelines for Quality Management in Projects. Available online: www.iso.org (accessed on 20 August 2021).
  14. ISO 9001:2015. 2005. Quality Management Systems. Requirements. Available online: www.iso.org (accessed on 16 February 2018).
  15. ISO 9004:2018. 2005. Managing for the Sustained Success of an Organization—A Quality Management Approach. Available online: www.iso.org (accessed on 29 January 2021).
  16. ISO/TR 10014:2021. 2021. Guidelines for Managing the Economics of Quality. Available online: www.iso.org (accessed on 20 August 2021).
  17. Jeong, Sukjae, Jaehyun Han, and Jihyun Kim. 2016. A Systematic Approach to Identify Promising New Items for Small to Medium Enterprises: A Case Study. Sustainability 8: 1158. [Google Scholar] [CrossRef] [Green Version]
  18. Jiang, Jiehui, and Kandachar Prabhu. 2008. New Market, New Challenge, New Opportunity (2)-User Context Research. Paper presented at the 30th Annual International IEEE EMBS Conference, Vancouver, BC, Canada, August 21–24; pp. 1583–86. [Google Scholar]
  19. Johannsen, Florian. 2013. A holistic approach for integrating methods in quality management. Paper presented at the 11th International Conference on Wirtschaftsinformatik (WI2013), Leipzig, Germany, February 27–March 1; vol. 2, pp. 999–1014. [Google Scholar]
  20. Khoshkangini, Reza, Peyman Sheikholharam Mashhadi, Peter Berck, Saeed Gholami Shahbandi, Sepideh Pashami, Sławomir Nowaczyk, and Tobias Niklasson. 2020. Early Prediction of Quality Issues in Automotive Modern Industry. Information 11: 354. [Google Scholar] [CrossRef]
  21. Lee, Pyoungsoo, and Yong Won Seo. 2017. Directions for Social Enterprise from an Efficiency Perspective. Sustainability 9: 1914. [Google Scholar] [CrossRef] [Green Version]
  22. Ling, Mei Goh, Sook Yeo Fern, Kah Lim Boon, and Seng Tan Huat. 2016. Understanding Customer Satisfaction of Internet Banking: A Case Study In Malacca. Procedia Economics and Finance 37: 80–85. [Google Scholar] [CrossRef] [Green Version]
  23. Liu, Weihua, Enze Bai, Liwei Liu, and Wanying Wei. 2017. A Framework of Sustainable Service Supply Chain Management: A Literature Review and Research Agenda. Sustainability 9: 421. [Google Scholar] [CrossRef] [Green Version]
  24. Malevskaia-Malevich, Ekaterina D., and A. Sergey Leonov. 2018. Factor Analysis Model of the Total Cost of Quality in the Enterprise Production Process. Paper presented at the 32nd Conference of the International Business Information Management Association (IBIMA), Seville, Spain, November 15–16; pp. 5886–90. [Google Scholar]
  25. Medne, Aija, and Inga Lapina. 2019. Sustainability and Continuous Improvement of Organization: Review of Process-Oriented Performance Indicators. Journal of Open Innovation Technology Market and Complexity 5: 49. [Google Scholar] [CrossRef] [Green Version]
  26. Mitchell, Mark L., and Janina M. Jolley. 2001. Research Design Explained, 4th ed. Beverly: Wadsworth Publisher. [Google Scholar]
  27. Mitreva, Elizabeta, Simona Gjurevska, Nako Taskov, and Hristijan Gjorshevski. 2019. Application of Methods and Techniques for Free Production in a Company for Manufacture of Bread and Bakery Products. Quality Access to Success 20: 141–45. [Google Scholar]
  28. Molina-Azorín, José F., Juan J. Tarí, Enrique Claver-Cortés, and Maria D. López-Gamero. 2009. Quality management, environmental management and firm performance: Areview of empirical studies and issues of integration. International Journal of Management Reviews 11: 197–222. [Google Scholar] [CrossRef]
  29. Munn, Pamela, and Eric Drever. 2004. Using Questionnaires in Small-Scale Research: A Beginner’s Guide. Glasgow: Scottish Council for Research in Education Center. [Google Scholar]
  30. Najwa, Abu Bakar, Selamat Ali, and Krejcar Ondrej. 2019. Improving Agent Quality in Dynamic Smart Cities by Implementing an Agent Quality Management Framework. Applied Sciences 9: 5111. [Google Scholar]
  31. Nenadál, Jaroslav, David Vykydal, and David Waloszek. 2018. Organizational Excellence: Approaches, Models and their use in Czech Organizations. Quality Innovation Prosperity 22: 47–64. [Google Scholar] [CrossRef] [Green Version]
  32. Nguyen, Minh Hue, Anh Chi Phan, and Yoshiki Matsui. 2018. Contribution of Quality Management Practices to Sustainability Performance of Vietnamese Firms. Sustainability 10: 375. [Google Scholar] [CrossRef] [Green Version]
  33. Okolnishnikova, Irina, Georgiy M. Greyz, Yulia G. Kuzmenko, and Viktor M. Katochkov. 2018. Selection of Logistics Performance Indicators. Paper presented at the 32nd Conference of the International Business Information Management Association (IBIMA), Seville, Spain, November 15–16; pp. 7395–400. [Google Scholar]
  34. Olaru, Marieta, Dorin Maier, Diana Nicoara, and Andreea Maier. 2014. Establishing the basis for Development of an Organization by Adopting the Integrated Management Systems: Comparative Study of Various Models and Concepts of Integration. Procedia—Social and Behavioral Sciences 109: 693–97. [Google Scholar] [CrossRef] [Green Version]
  35. Parameshwaran, Rajagopalan, and Parthi S. Srinivasan. 2008. An Integrated Closed-loop Model for Service Performance Management. International Journal of Services and Operations Management 4: 34–55. [Google Scholar] [CrossRef]
  36. Popescu, Dan, Antoni Garcia-Sanchez, Valentina Nicolae, Cristina State, and Livia Toanca. 2015. Empirical Research Regarding the Importance of Volunteering for the Sustainable Development of Social Economy Enterprises. Economic Computation and Economic Cybernetics Studies and Research 49: 49–67. [Google Scholar]
  37. Proksch, Dorian, Julia Busch-Casler, Marcus Max Haberstroh, and Andreas Pinkwart. 2019. National health innovation systems: Clustering the OECD countries by innovative output in healthcare using a multi indicator approach. Research Policy 48: 169–79. [Google Scholar] [CrossRef]
  38. Roth, Aleda V., and Larry J. Manor. 2009. Insights into service operations management: A research agenda. Production and Opererations Management 12: 145–64. [Google Scholar] [CrossRef]
  39. Schaltegger, Stefan, and Marcus Wagner. 2006. Integrative management of sustainability performance, measurement and reporting. International Journal of Accounting, Auditing and Performance Evaluation (IJAAPE) 3: 1–19. [Google Scholar] [CrossRef]
  40. Schwab, Donald P. 2005. Research Methods for Organizational Studies, 2nd ed. New York: Psychology Press. [Google Scholar]
  41. Siva, Vanajah, Ida Gremyr, Bjarne Bergquist, Rickard Garvare, Thomas Zobel, and Raine Isaksson. 2016. The support of Quality Management to sustainable development: A literature review. Journal of Cleanner Production 138, Pt 2: 148–57. [Google Scholar]
  42. Spector, Paul E. 2012. Industrial and Organizatiional Psychology. Research and Practice, 6th ed. Singapore: John Wiley and Sons. [Google Scholar]
  43. Staš, David, Radim Lenort, Pavel Wicher, and David Holman. 2015. Green Transport Balanced Scorecard Model with Analytic Network Process Support. Sustainability 7: 15243–61. [Google Scholar] [CrossRef] [Green Version]
  44. Tigist, Fetene Adane, and Mihai Nicolescu. 2018. Towards a Generic Framework for the Performance Evaluation of Manufacturing Strategy: An Innovative Approach. Journal of Manufacturing and Materials Processing 2: 23. [Google Scholar]
  45. To, Wai Ming, Billy TW Yu, and Peter KC Lee. 2018. How Quality Management System Components Lead to Improvement in Service Organizations: A System Practitioner Perspective. Administrative Sciences 8: 73. [Google Scholar] [CrossRef] [Green Version]
  46. Wagner, I. 2021. Automotive Industry Worldwide—Statistics & Facts. Available online: https://www.statista.com/topics/1487/automotive-industry/ (accessed on 20 August 2021).
  47. Werner-Lewandowska, Karolina, and Paulina Golinska-Dawson. 2021. Sustainable Logistics Management Maturity—The Theoretical Assessment Framework and Empirical Results from Poland. Sustainability 13: 5102. [Google Scholar] [CrossRef]
  48. Wismantoro, Yohan, Herry Subagyo, and Vincent Didiek Wiet Aryanto. 2019. Model for Website Service Quality and Technology Acceptance Factors: A Case Study of Online Shoppers in Indonesia. Quality Access to Success 20: 89–94. [Google Scholar]
  49. Yee, Rachel WY., Peter KC. Lee, Andy CL Yeung, and TC Edwin Cheng. 2013. The relationships among leadership, goal orientation, and service quality in high-contact service industries: An empirical study. International Journal of Production Economics 141: 452–64. [Google Scholar] [CrossRef]
  50. Young, Sik Cho, Joo Yung Y. Jung, and Kevin Linderman. 2017. The QM evolution: Behavioral quality management as a firm’s strategic resource. International Journal of Production Economics 191: 233–49. [Google Scholar]
Figure 1. Concrete steps in the elaboration of the study (developed by the author).
Figure 1. Concrete steps in the elaboration of the study (developed by the author).
Jrfm 14 00480 g001
Figure 2. Stages of designing questionnaires (Jiang and Prabhu 2008, Figure 4, p. 1584).
Figure 2. Stages of designing questionnaires (Jiang and Prabhu 2008, Figure 4, p. 1584).
Jrfm 14 00480 g002
Figure 3. Links between the analyzed criteria.
Figure 3. Links between the analyzed criteria.
Jrfm 14 00480 g003
Figure 4. Graphical analysis for the distribution of the questionnaire.
Figure 4. Graphical analysis for the distribution of the questionnaire.
Jrfm 14 00480 g004
Figure 5. (a) Global scores obtained by each enterprise; (b) Global scores obtained for each criterion.
Figure 5. (a) Global scores obtained by each enterprise; (b) Global scores obtained for each criterion.
Jrfm 14 00480 g005
Figure 6. Total scores obtained in each section by the enterprises.
Figure 6. Total scores obtained in each section by the enterprises.
Jrfm 14 00480 g006
Figure 7. (a) Competitor 1 individual scores; (b) Competitor 2 individual scores; (c) Competitor 3 individual scores; (d) Competitor 4 individual scores; (e) Competitor 5 individual scores; (f) Competitor 6 individual scores.
Figure 7. (a) Competitor 1 individual scores; (b) Competitor 2 individual scores; (c) Competitor 3 individual scores; (d) Competitor 4 individual scores; (e) Competitor 5 individual scores; (f) Competitor 6 individual scores.
Jrfm 14 00480 g007
Figure 8. Frequency of occurrence by categories of difficulty.
Figure 8. Frequency of occurrence by categories of difficulty.
Jrfm 14 00480 g008
Figure 9. Answers regarding the management criteria.
Figure 9. Answers regarding the management criteria.
Jrfm 14 00480 g009
Figure 10. Answers regarding the strategy criteria.
Figure 10. Answers regarding the strategy criteria.
Jrfm 14 00480 g010
Figure 11. Answers regarding the management systems criteria.
Figure 11. Answers regarding the management systems criteria.
Jrfm 14 00480 g011
Figure 12. Answers regarding the customer management criteria.
Figure 12. Answers regarding the customer management criteria.
Jrfm 14 00480 g012
Figure 13. Answers regarding the employee management criteria.
Figure 13. Answers regarding the employee management criteria.
Jrfm 14 00480 g013
Figure 14. Answers regarding the risk management criteria.
Figure 14. Answers regarding the risk management criteria.
Jrfm 14 00480 g014
Figure 15. Answers regarding the performance and results obtained criteria.
Figure 15. Answers regarding the performance and results obtained criteria.
Jrfm 14 00480 g015
Table 1. Evaluation rules on Sections 1–6 of questionnaire.
Table 1. Evaluation rules on Sections 1–6 of questionnaire.
Section (S)Respondent AnswerAssessmentScores
S1; S2;
S3: 3.3 to 3.5;
S4; S5;
S6: 6.1; 6.3 to 6.5.
NoWe do not perform this activity0
PartialThere is a plan for carrying out this activity3
YesWe are implementing this activity5
S3: 3.1ISO 9001Reference standard10
Other documentsOther quality requirements5
Not applicableWe are not implementing this activity0
S3: 3.2NoWe are not implementing this activity0
Yes, ISO 9001 + 1 systemISO 9001 + 1 system5
Yes, ISO 9001 + 2 systemsISO 9001 + 2 systems8
Yes, ISO 9001 + 3 systems or more than 3ISO 9001 + 3 systems or more than 310
S6: 6.2NoWe are not implementing this activity0
When changes are made to the premises that formed the basis for their establishmentWe perform checks on the premises3
YesThere is a well-established plan for carrying out this activity5
Table 2. Evaluation rules on Section 7 of questionnaire.
Table 2. Evaluation rules on Section 7 of questionnaire.
Section (S)Respondent AnswerAssessmentScores
S7NoThe results are not measured, or the objectives have not been met0
PartialMost of the objectives have been met3
YesThe objectives have been met or even advanced5
Table 3. Maximum score that can be obtained.
Table 3. Maximum score that can be obtained.
S1S2S3S4S5S6S7
Total max/criteria25253525252525
Total max/questionnaire185
Table 4. Questionnaire analysis.
Table 4. Questionnaire analysis.
Target GroupNumber of Questionnaires SentReceived AnswersValidated AnswersCanceled Answers *Answers Received (%)Validated Answers (%)Canceled Answers * (%)
Managers107617085.7114.28
* The cancellation was due to the absence/duplication of an answer to at least one of the questions.
Table 5. Vehicle manufacturers.
Table 5. Vehicle manufacturers.
No.ManufacturersUnits Sold
2018201720162015
1Dacia51,24229,97427,59726,840
2Group VW14,601936183397184
3Renault12,523770869104900
4Ford9503804368345749
5Toyota5179375230892863
6Mazda2259189716531467
Table 6. Romanian market companies operating in the field of car services.
Table 6. Romanian market companies operating in the field of car services.
No.ManufacturersTurnover (RON)Strategies Related to Performance Management
20182017
1Dacia29,496,295.027,573,421.0Customer Satisfaction, Quality, Process Improvement
2Group VW158,202,631.0151,591,646.0Environment, Customer Orientation, Compliance
3Renault188,319,360.0154,096,604.0Efficiency, Quality, Standardization, Environment
4Ford177,259,928.0148,459,972.0Human Resource Development, Clients, Quality, Eco, Social Responsibility
5Toyota152,051,923.012,785,133.0Continuous Improvement, Customers, Employees, Risk, Safety, Environmental Quality, Efficiency
6Mazda733,006,910.070,358,697.0Customer Satisfaction, Quality Environment Security, Processes, Technology
Table 7. Answers collected regarding the questions related to Hypothesis H1.
Table 7. Answers collected regarding the questions related to Hypothesis H1.
Unit/SectionC1C2C3C4C5C6Total
S1192021252520130
S2182120182520122
S3103131313530168
S4212125212521134
S5111914212319107
S6232323192323134
S7132113182118104
Total115156147153177151
Table 8. Percentage of achievement for the declared criteria.
Table 8. Percentage of achievement for the declared criteria.
NoStrategies Related to Performance ManagementPercentage of Achievement of the Criteria (%)
1Customer satisfaction84%
Quality28.6%
Process improvement52%
2Environment88.6%
Customer orientation84%
Conformity84%
3Efficiency52%
Quality—Environment88.6%
Standardization80%
4Human resource development84%
Clients84%
Quality Eco88.6%
Social Responsibility72%
5Continuous improvement100%
Customers100%
Employees92%
Risk and Safety92%
Quality and Environment84%
6Customer satisfaction85.7%
Quality—Environment72%
Security Processes84%
Technology80%
Global results81%
Table 9. Answers collected regarding the questions related to Hypothesis H2.
Table 9. Answers collected regarding the questions related to Hypothesis H2.
NoCategories of ProblemsNo. of AppearancesFrequency (%)
1Insufficient employee involvement211.11
2Lack of indicators on employee satisfaction211.11
3Negative feedback received from customers316.67
4Lack of integrated information systems15.556
5Employee motivation tools are limited15.556
6Large staff turnover422.22
7Long duration of solving service interventions15.556
8Lack of a program for the integration of ethnic minorities15.556
9Asynchronization of expenses with receipts211.11
10Lack of a program to attract young high school/college graduates15.556
Total:8100
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Lascu, E.; Severin, I.; Lascu, F.D.; Gudana, R.A.; Nalbitoru, G.; Ignat, N.D. Framework on Performance Management in Automotive Industry: A Case Study. J. Risk Financial Manag. 2021, 14, 480. https://doi.org/10.3390/jrfm14100480

AMA Style

Lascu E, Severin I, Lascu FD, Gudana RA, Nalbitoru G, Ignat ND. Framework on Performance Management in Automotive Industry: A Case Study. Journal of Risk and Financial Management. 2021; 14(10):480. https://doi.org/10.3390/jrfm14100480

Chicago/Turabian Style

Lascu, Elena, Irina Severin, Florina Daniela Lascu, Razvan Adrian Gudana, Gabriela Nalbitoru, and Nicoleta Daniela Ignat. 2021. "Framework on Performance Management in Automotive Industry: A Case Study" Journal of Risk and Financial Management 14, no. 10: 480. https://doi.org/10.3390/jrfm14100480

APA Style

Lascu, E., Severin, I., Lascu, F. D., Gudana, R. A., Nalbitoru, G., & Ignat, N. D. (2021). Framework on Performance Management in Automotive Industry: A Case Study. Journal of Risk and Financial Management, 14(10), 480. https://doi.org/10.3390/jrfm14100480

Article Metrics

Back to TopTop