Process and Product Change Management as a Predictor and Innovative Solution for Company Performance: A Case Study on the Optimization Process in the Automotive Industry

Abstract

Change and innovation are increasingly exerting a significant influence on the daily activities of companies. To ensure optimal control, innovative solutions are employed that are encapsulated in the concept of change management. In the engineering change sector, the proposed approach involves developing solutions and making continuous adjustments to the manufacturing process to enhance productivity and to meet customer needs. Within the automotive industry, companies utilize innovations and process change management to continuously improve and strengthen their position in the market, such as KPI/KPRS and PCI. To achieve this, the present study gathers real digital data from the Romanian branches of two renowned automotive companies. The data regarding change requests include 215 registrations for the first company and 734 registrations for the second company. By employing complex statistical methods such as ANOVA, Student’s t-test, the Mann–Whitney test, and a regression model, the primary objective of this study is to model and to identify the best predictor of change request status. Additionally, this study aims to explore how this change process influences the economic performances of the companies and the performance indicators of change management in manufacturing processes. The findings indicate that, both in the organizations in general and within the automotive industry, when products experience high demand in the market, the number of change requests increases. This highlights the importance of internal optimization of the automation system. Moreover, the study results underscore the crucial role of an effective smart manufacturing and optimal change management system to uphold and to enhance the economic performance of automotive companies.

1. Introduction

Change management is increasingly becoming a crucial aspect of company daily operations. To ensure optimal control, special management techniques are employed that focus on the human factor, because the support of employees is vital for successful change implementation [1]. From any point of view, the most important factor is the human factor, because the implementation of change actively depends on the support of employees [1]; for example, one of the well-known methods is the people-centered change implement model (PCI^®) [2]). The PCI^® method is a comprehensive process that provides information to empower change managers to implement innovations in organizations in a customized and effective way [3] The method focuses on the topics in significant organizational change processes or initiatives. PCI^® provides a systematic approach to planning, executing, and controlling all personal aspects of the change process [4]. The PCI^® method is based on the development of management skills, tools, and processes for change planning and implementation [5]. Concurrently, development processes are acknowledged as pivotal for innovations in physical products. Research suggests that an effective change management strategy consists of three key phases: pre-change preparation, change implementation, and assessing user impact [6].

The inevitability of change in original product design is crucial for improving quality, productivity, and customer satisfaction in the automotive manufacturing environment. Successfully predicting the impact of changes in product design is a key factor in the engineering change management (ECM) process. Various change propagation and impact analysis methods have been proposed by researchers, with a primary focus on product architecture and the selection of change propagation paths during the design stage of the product life cycle. However, these tools have limitations as they mainly aim to identify the negative impacts of changes. Based on case studies and past data interpretation, several challenges related to ECM have been identified [7] which include: (1) Difficulty in effectively communicating changes across the supply chain and to all stakeholders. (2) The need for early triggers to prompt action on engineering change requests (ECRs) with anticipated task failures. (3) The necessity for improved capabilities of ERP/inventory management software to effectively handle change.

Despite the recognized benefits, the industry has not fully embraced an integrated digital support system for mass customization to manage ECM efficiently. There is a gap in understanding and estimating the impact of changes in the manufacturing environment. To address these challenges and to enhance change management practices, there is a significant need for the development of change prediction software. Such software would allow for a comprehensive understanding of the impacts of changes well in advance, enabling better planning for change implementation throughout the supply chain and incorporating necessary controls. This advancement would significantly improve the ECM process and contribute to the overall success and efficiency of automotive manufacturing operations [7].

For successful and sustainable change, effective coordination is necessary. Thus, an integrative change management model is needed that reflects its dimensions and the requirements that become essential [8]. Change management fosters individual agility and adaptability, while it aligns an organization for effectiveness. Pioneering studies by Lester Coch and John French, in 1948, explored motivational issues related to change initiatives and identified behavioral resistance [9]. Their scientific contributions consisted of defining the three types of change: planned, continuous, and transitory [10]. As the 1990s saw globalization and rapid technological advancements, managing change became critical for business success, as exemplified by John P. Kotter’s eight-step model [11] and Kurt Lewin’s three-phase model [12].

Recently, the literature has suggested that resistance to change can be constructive and should not be disregarded in engineering change contexts. Instead of viewing resistance as an opponent, classical management theory now emphasizes carefully managing resistance for optimal results [13].

Another study introduced an engineering change management (ECM) decision support solution and utilized a five-step process for preprocessing, modeling, enhancing model performance, and providing interpretable results to engineers. The proposed solution offered an early assessment of change consequences for industries, aiding engineers in decision-making regarding change impact, lead time, and whether to proceed with changes that have significant effects on other systems and require extensive evaluation. The industrial case study demonstrated promising performance, and validation by domain experts confirmed the practical applicability, usefulness, and logical correctness of the approach [14]. Compared to existing research, this solution stands out for its full automation, making it convenient for engineers. Additionally, it is one of the rare data-driven ECM solutions capable of handling a mixed format of attributes, including text, categorical, and numerical data. The inclusion of text data provides valuable insights.

However, the above-mentioned work has some limitations. Data quality, particularly imbalanced datasets and issues with text data, could have affected the solution’s performance. Future work should consider expanding the scope of change impacts to include system requirements, functions, change costs, and suggestions for redesign. The prediction of lead time could be enhanced by considering the critical level of changes. Moreover, it is crucial to consider current circumstances and technological advancements when applying the solution to new components and functions. In conclusion, the tested ECM decision support solution aims to enhance the quality, efficiency, and transparency of ECM processes, thus, reducing manual work in product development and optimizing time-to-market for complex products [14].

Engineering change has steadily grown in importance both as an important issue for industry and as an active area of academic research. Change is defined as a change made to parts/products, designs, or software that have already been released into production and are already in a lifecycle. A change may include any intervention in the form and/or function of the product, in whole or in part, and may alter the interactions and dependencies of the product’s constituent elements. Key aspects of the engineering change process have been highlighted along with the tools and methods available to support the process. Important related areas such as organizational structure and employee skills have also been highlighted [15]. Some documents have examined the principles of change, effective implementation methods within an organization, the significance of change management, and the role of management in facilitating change [16]. Engineering change plays a pivotal role in product or process innovation and engineering change management has gained prominence in research and the automotive industry. Managing engineering change necessitates balancing the efficiency and effectiveness of the change management system, and requires guidelines to ensure accurate change requests [17]. Some studies have indicated that change demand can be detrimental, especially when products have longer lead times [18]. Another referential research introduced a methodology based on the Value Stream Mapping (VSM) 4.0 approach for continuous improvement in the downstream supply chain distribution process. The study included an analysis of the current state through VSM mapping and the design of a future state, optimizing competitive tasks, considering the country of origin and integrating Industry 4.0 technologies accepted by the company in terms of implementation time and budget. The use of the Lean Value Stream Mapping 4.0 approach enables the analysis and optimization of product and information flow throughout the value chain from suppliers to customers. Real-time monitoring of value streams through Industry 4.0 technologies allows quick resolution of potential waste, creating a fully integrated logistics environment with real-time data transmission between warehouses and customers. The results demonstrate improvements in economic, social, and environmental performance. While the integration of new technologies requires a significant initial investment, the benefits gained from the improvements quickly justify the costs [19].

Changes in production systems offer insights into ongoing developments, but errors in implementation must be addressed to achieve quality objectives [18,20,21]. Furthermore, change requests significantly influence the bill of materials (BOM), directly related to production [22]. The purpose of this study is an in-depth analysis of the specialized literature complemented by applied research within two Romanian companies active in the automotive industry, subsidiaries of globally recognized companies worldwide (details in

Table 1. General information about the companies.

Characteristics	Company A	Company B
The field of activity for Romanian branch	Mobility Solutions: Chassis Control Systems and Propulsion Systems (CC–PS)	The manufacture of electrical and electronic equipment for motor vehicles and motor vehicle engines
The number of employees (for 2021)	3040	3682
Head office	Stuttgart (Germany)	Hanover (Germany)
The type of certification	ISO 9001	ISO 14001
Start date of activity in Romania	19 September 2006	10 June 2005
The mean number of yearly changes	70	240
Turnover (for 2021)—in million lei	2,719,582.6	3,361,428.4

(Source: Prepared by the authors based on the information from www.listafirme.ro. Accessed on 30 January 2023) [23].

), to identify the main problems and to highlight that there is a strong and clear industrial need for support in the engineering change process, especially with the identification of potential change propagation paths. The automotive industry is in a constant state of development, and optimization activities are increasingly present in the product development process. Another reason for choosing the two companies is the fact that the research is part of the doctoral research of the first author and also has important and direct contributions in the change management department of the companies.

This study aims to conduct a comprehensive analysis of the specialized literature and to apply research within two Romanian companies in the automotive industry, to reveal the importance of support in the engineering change process, particularly by identifying potential change propagation paths. The conceptual framework (Figure 1) is based on formal change procedures [24] and Terwiesch and Loch’s four principles of change management [25]:

• Avoiding inappropriate changes;
• Reducing the negative effects of change demands;
• Early detection of engineering changes;
• Streamlining the urgent change request process.

The proposed approach focuses on innovations within the engineering change sector that can be developed through various methods, including adapting the production flow/process and generating new ideas, products, services, and practices. The primary goal is to enhance productivity by efficiently utilizing available resources and addressing evolving customer demands. It goes beyond merely fulfilling existing needs; instead, it involves creating new demands by offering products or services that customers may not have explicitly requested but would find appealing and satisfying. In this way, automotive companies establish competitive advantages by emphasizing innovative processes to bolster their market position. Every change request presents an opportunity for improvement.

The impacts of change requests are directly observable in the final manufactured product or service that a company delivers to its customers.

2. Materials and Methods

For this case study, data were collected from two important companies from Romania, named Company A and Company B to protect their names/brands, as follows:

• For company A, there were 215 registrations between April 2018 and December 2020;
• For company B, there were 734 registrations between July 2020 and January 2023.
  In addition, economic indicators were collected from a national internet page (www.listafirme.ro) (accessed on 30 January 2023) [23].
  The variables used in the study are both categorical and continuous, mentioning for each categorical variable the codification used by the authors for the statistical software SPSS 23.0 (this step is necessary for statistical analysis of all the variables as nominal and/or ordinal express in words and to facilitate the applications of statistical methods through the specific software) as follows:
• Categorical variables (expressed by words, namely also qualitative/“dummy” variables):

  ○

  The month of proposal, the planned month for implementing, and the effective implementation month of change requests were assigned the following SPSS codes: 1, January; 2, February; …; 12, December.

  ○

  The year of proposal, the year for implementing, and the year of effective implementation of change requests were assigned the following SPSS codes: 1, 2018; 2, 2019; 3, 2020; 4, 2021; 5, 2022; 6, 2023.

  ○

  The change request statuses were assigned the following SPSS codes: 10, preparation; 11, canceled; 20, manager approval; 22, planning phase; 24, validation phase; 30, start implementation; 40, local confirmation for implementation; 44, global confirmation for implementation; 48, closed.

  ○

  The types of processes proposed for change were assigned the following SPSS codes: 1, engineering change and 2, correction.

  ○

  The number of days from creation to planning of change requests and the number of days from planning to effective implementation of change requests were assigned the following SPSS codes: 1, <10 days; 2, 11–20 days; 3, 21–30 days; 4, 31–50 days; 5, 51–100 days; 6, 101–200 days; 7, 201–300 days; 8, >300 days.

  ○

  Customer involvement in a change request was assigned the following SPSS codes: 1, yes or 2, no.

• Continuous variables (expressed by numbers, namely also quantitative/numerical variables) for all the economic indicators for the last five years, i.e., 2017–2021, were as follows: turnover, net profit, own capital, number of employees, debts, and fixed assets (all of these indicators in Romanian lei).

Table 1 presents some details regarding the companies included in the research.

Table 2. Performance indicators for the two companies (in Romanian lei current prices).

	Year	2017	2018	2019	2020	2021
Performance Indicators
Company A
Turnover		1,772,494,278	2,245,269,652	2,642,519,314	2,384,450,466	2,719,582,643
Net profit		75,536,047	131,317,375	104,699,652	5,525,460	67,059,975
Own capital		278,241,373	407,843,613	514,112,683	518,776,213	585,836,188
The number of employees		2718	3209	3269	3006	3040
Debts		719,133,992	790,444,495	757,731,299	772,423,357	733,595,299
Fixed assets		619,127,854	749,578,670	789,028,236	723,823,357	693,030,091
Company B
Turnover		3,769,456,584	4,292,798,906	4,097,920,130	3,549,865,965	3,361,428,353
Net profit		59,103,202	852,506,056	134,188,496	225,919,057	−244,172,664
Own capital		990,543,637	1,849,238,824	1,983,427,320	2,209,346,377	1,976,488,993
Number of employees		3341	3881	3890	3718	3682
Debts		1,159,496,262	1,001,313,179	915,120,917	1,034,944,775	1,630,408,279
Fixed assets		1,619,664,158	2,010,338,748	2,168,836,425	2,320,589,143	2,667,134,091

(Source: Prepared by the authors based on the information from [23] www.listafirme.ro. Accessed on 30 January 2023).

lists the economic indicators of performance for the last five years for each company and practically all the continuous/numerical variables from the study are presented for a better visualization of similarities and differences of these companies. All the data were transformed into current prices and are expressed in Romanian lei, for a correct dynamic analysis from an economic point of view.

Because this research is based on big data collected from two automotive companies, complex statistical methods were applied according to the type of data, i.e., qualitative/categorical and/or continuous/numerical data. First, descriptive statistics were calculated by using the statistical software SPSS 23.0, then the normal distribution of the numerical data was tested with the aim to choose which statistical method would be applied. In this section, we present a short description of the chosen and applied statistical methods.

Descriptive statistics were calculated for each economic indicator, and the results were presented classically as mean ± standard deviation (minimum value − maximum value) for each continuous variable mentioned in Table 2 and as absolute and relative frequencies for categorical/“dummy” variables, separately for each company.

To test if there were statistically significant differences between whether or not customers are implicated in the change process with respect to change request status, a chi-square bivariate test was applied.

The Student t-test for independent samples was applied to test if there were statistically significant differences between the two companies with respect to the following:

• The average number of days from creation/initiation to planning of change requests.
• The average number of days from planning to effective implementation of change requests.

An ANOVA test was applied to test and to verify if there were:

• Statistically significant differences (only for Company A these data were collected being available) according to whether or not customers are implicated in the change process, referring to the average number of days from creation/initiation to planning of change request and the average number of days from planning to effective implementation of a change request.
• In addition, although both companies are active in the automotive industry and have some similarities, the ANOVA test was applied to verify if the data have (or not) important variations around mean values, and whether or not these variations are statistically significant between companies for all the economic performance indicators.

Statistically significant differences between the mean values of the two companies refer to all the economic efficiency indicators calculated per employee (i.e., the turnover, net profit, and debts). To identify and verify if there are statistically significant differences between Company A and Company B, regarding the month of creation/initiation of change requests, the month planned for implementation, the month of effective implementation of change requests, and the change request status, we applied the Mann—Whitney U test for independent samples. This statistical test allows us to test if the distribution of each “dummy”/categorical variable is the same along each of the companies or if there are statistically significant differences. Boxplots are used as graphical representations for better visualization of the quantitative differences between two companies from the research detailed for each type of process proposed for change.

Since the main purpose of this research was to find the best predictor of change request status (considered as the dependent variable from a statistically point of view in this research), a multilinear regression analysis with the Enter method was applied by using the same statistical software, i.e., SPSS 23.0. The following independent variables of the regression model were considered: the month of proposal for a change request, the planned month for implementing a change request, the month for effective implementation of a change request, type of processes proposed for changing, the number of days from creation to planning of change requests, and the number of days from planning to effective implementation of change requests. This statistical method allowed us to also quantify the contribution (positive or negative) of the modification with one unit of each independent variable on the dependent variable (change request status), to identify if it was important for the change management process, including the month of proposal, the month of request, the month of effective implementation, the type of process, and the number of days between the proposal phase and the request and the implementation phases.

For all statistical analyses, the SPSS 23.0 software was used. For statistical significance, a threshold of p-value <0.05 was considered for the inference methods and p-value <0.1 for the regression analysis. Microsoft Excel was used for some graphical representations. All the results are presented in detail in the next section, most of them comparatively, between Company A and Company B.

3. Results

The first results of this research refer to descriptive statistics for each company and each variable from the study.

Table 3. Descriptive statistics for the economic performance indicators (in lei current prices).

Indicators	Company A	Company B
Turnover (in Romanian lei)	2,352,863,270.6 ± 376,757,888.7 (1,772,494,278–2,719,582,643)	3,814,293,987.6 ± 382,893,784 (3,361,428,353–4,292,798,906)
Net profit (in Romanian lei)	76,827,701.8 ± 47,241,551.19 ((5,525,460–131,317,375)	205,508,829.4 ± 402,416,647 (−244,172,664–852,506,056)
Own capital (in Romanian lei)	460,962,014 ± 120,384,209.6 (278,241,373–585,836,188)	1,801,809,030.2 ± 471,695,374 (990,543,637–2,209,346,377)
The number of employees	3,048.4 ± 215.342 (2718–3269)	3,702.4 ± 222.689 (3341–3890)
Debts (in Romanian lei)	754,665,688.4 ± 28,772,028.62 (719,133,992–790444495)	1,148,256,682.4 ± 283,452,270 (915,120,917–1,630,408,279)
Fixed assets (in Romanian lei)	714,917,641.6 ± 64,085,542.71 (619,127,854–789,028,236)	2,157,312,513 ± 386,450,883 (1,619,664,158–2,667,134,091)
Turnover per employee (in Romanian lei)	769,599.55 ± 95,358.699 (652132−894600)	1,031,102.43 ± 94,007.845 (912,935–1,128,242)
Net profit per employee (in Romanian lei)	199,435.67 ± 388,880.289 (1838–894,600)	53,259.17 ± 104,476.410 (−66,315–219,661)
Debts per employee (in Romanian lei)	248,194.19 ± 12,896.237 (231,793–264,582)	312,293.97 ± 84,077.189 (235,250–442,805)

presents the descriptive statistics (presented as mean ± standard deviation (minimum value − maximum value) in Romanian lei current prices) for the economic performance indicators, including the yield indicators for each company.

Since there are some differences between the companies, as shown in Table 1 and Table 2, we calculated the efficiency per employee, which is shown in Table 3, to obtain a better comparison of each company’s performance. The differences are obvious. The turnover of company A is under the value of Company B per employee, and the net profit per employee is superior. In addition, debts per employee for Company A is lower than that for Company B.

In Figure 2a–f, the evolutions and trends of the main economic indicators of the researched companies are comparatively described.

It can be seen that the evolutions of the companies are different during the five years, except for two indicators, i.e., own capital (with a steep trend for Company B and a smooth trend for Company A) and the number of employees, both with an ascending trend. For example, the turnover for Company B has a descending trend compared with that of Company A, which has an ascending trend. More interestingly, the evolution of net profit has a positive trend for Company A and a negative trend for Company B, while debts and fixed assets have opposite evolutions, i.e., for Company B debts and fixed assets increase, while for Company A they decrease with positive significance, of course, for Company A by maintaining the good evolution for this company.

Starting with the above-mentioned results, we applied an ANOVA test to compare if the means were statistically different between the companies; the results are presented in

Table 4. ANOVA results, in means, for the economic indicators.

		Sum of Squares *	df	Mean Square *	F	Sig.
Turnover	Between groups	5,339,449,351,477.000	1	5,339,449,351,477.000	37.008	0.000
	Within groups	1,154,216,627,304.000	8	144,277,078,413.000
	Total	6,493,665,978,782.000	9
Net profit	Between groups	403,655,124,023.000	1	403,655,124,023.000	0.518	0.492
	Within groups	6,233,566,658,353.000	8	779,195,832,294.000
	Total	6,637,221,782,376.000	9
Own capital	Between groups	4,494,676,802,131.000	1	44,94,676,802,131.000	37.932	0.000
	Within groups	947,955,536,899.000	8	118,494,442,112.000
	Total	54,426,323,390,310.000	9
The number of employees	Between groups	1.000	1	1.000	22.286	0.002
	Within groups	0.383850400	8	0.0.00
	Total	1.400	9
Debts	Between groups	387,284,676,394.000	1	387,284,676,394.000	9.542	0.015
	Within groups	324,692,075,434.000	8	40,586,509,429.000
	Total	711,976,751,829.000	9
Fixed assets	Between groups	5,201,257,412,602.000	1	5,201,257,412,602.000	67.790	0.000
	Within groups	61,380,4968,217.000	8	76,725,621,027.000
	Total	5,815,062,380,819.000	9
Turnover/employee	Between groups	170,959.244	1	17,095.244	19.069	0.002
	Within groups	71,723.404	8	896.176
	Total	242,682.648	9
Net profit/employee	Between groups	53,418.956	1	53,418.956	0.659	0.440
	Within groups	648,572.978	8	81,071.622
	Total	7,019,912.934	9
Debts/employee	Between groups	10,271.798	1	10,271.798	2.839	0.130
	Within groups	28,941.245	8	3617.906
	Total	39,213.043	9

(Note: * in million lei.)

.

The results of the ANOVA test indicate statistical differences between means (p-value <0.05) for the following: turnover, own capital, number of employees, debts, and fixed assets. Regarding the performance per employee, only turnover per employee showed significant differences according to the results from Table 4. These results practically confirm the differences observed in Figure 2b and the means for net profit per employee and debts per employee from Table 3. After this economic performance description and analysis of the researched companies, we present the results referring to change processes in these companies. Therefore, in

Table 5. Descriptive statistics for the month of initiation, the planned month, and the month for effective implementation of change requests.

Month	The Month of Proposal/Initiation of Change Request				Planned Month for Implementing Change Request				The Month for Effective Implementation Month of Change Request
	Company A		Company B		Company A		Company B		Company A		Company B
		%		%		%		%		%		%
January	16	7.4	72	9.8	23	10.7	75	10.2	11	5.1	61	8.3
February	19	8.8	73	9.9	14	6.5	64	8.7	7	3.3	75	10.2
March	20	9.3	72	9.8	14	6.5	70	9.5	8	3.7	77	10.5
April	19	8.8	40	5.4	16	7.4	46	6.3	8	3.7	37	5.0
May	15	7.0	53	7.2	26	12.1	55	7.5	12	5.6	62	8.4
June	20	9.3	63	8.6	14	6.5	57	7.8	15	7.0	52	7.1
July	28	13.0	75	10.2	27	12.6	76	10.4	26	12.1	72	9.8
August	16	7.4	44	6.0	24	11.2	50	6.8	16	7.4	54	7.4
September	16	7.4	53	7.2	9	4.2	54	7.4	27	12.6	50	6.8
October	19	8.8	68	9.3	14	6.5	65	8.9	24	11.2	64	8.7
November	12	5.6	54	7.4	11	5.1	55	7.5	36	16.7	58	7.9
December	15	7.0	67	9.1	23	10.7	66	9.0	25	11.6	71	9.7
Total	215	100.0	734	100.0	215	100.0	734	100.0	215	100.0	734	100.0

, we present the absolute and relative frequencies for the month of proposal/initiation of change requests, the planned month for implementing change requests, and the month of effective implementation of change requests.

Figure 3a–c show, comparatively, the two types of processes proposed for change (data clustered by company).

For the month of proposal for change requests (Figure 3a) and the planned month for implementation (Figure 3b), there are no differences between the companies for the process types, i.e., engineering change or correction. Regarding the month of effective implementation of change requests, there are visible differences between the companies with respect to both types of processes.

Regarding the year of proposal/initiation, the planned year for implementation, and the year of effective implementation of change requests, the data are presented in Figure 4, comparatively for the two companies. It can be seen that:

• For company A, in 2018 from a total of 51 proposals, 45% of the change requests were planned and 14% of the change requests were implemented; the situation is much better in 2019 and in 2020, by increasing the number of effectively implemented change requests and including the previous change requests from 2018.
• For company B, the situation is more equilibrated, although the number of change requests is superior compared with Company A.

In Figure 5 for Company A and in Figure 6 for Company B, we present the structures of the change requests’ statuses.

The situations are quite different: For Company A, most of the change requests are closed (75%) followed by local confirmation for implementation (8%) and canceled (8%); for company B, only 10% of the change requests are closed but 86% of the change requests receive local confirmation for implementation and 2% of the change requests already have manager approval.

The structures of the types of processes proposed for change are presented in Figure 7. For Company A, the number of engineering changes is relatively equal with the corrections, but the structure is different for Company B with 66.3% correction and 33.7% engineering changes.

Regarding the number of days from creation to planning of a change request and the number of days from planning to effective implementation of a change request, in Figure 8a,b we present the boxplots clustered by company for each type of process change request; the differences between the two researched companies are visible.

The detailed structures of the indicators are presented in

Table 6. Descriptive statistics for the number of days between the phases (i.e., proposal, planning, and implementation phases).

Intervals	The Number of Days from Creation to Planning of Change Requests				The Number of Days from Planning to Effective Implementation of Change Requests
	Company A		Company B		Company A		Company B
		%		%		%		%
<10 days	10	4.7	422	57.5	78	36.3	5	0.7
11–20 days	23	10.7	196	26.7	11	5.1	371	50.5
21–30 days	10	4.7	72	9.8	11	5.1	253	34.5
31–50 days	21	9.8	29	4.0	11	5.1	82	11.2
51–100 days	34	15.8	10	1.4	29	13.5	18	2.5
101–200 days	54	25.1	2	0.3	42	19.5	2	0.3
201–300 days	32	14.9	0	0	33	15.3	1	0.1
>300 days	31	14.4	2	0.3	78	36.3	2	0.3
Total	215	100.0	734	100.0	215	100.0	734	100.0

as follows:

• Regarding the number of days from creation to planning of a change request, for company A, the numbers of days between creation and planning are mostly (25%) approximately 101–200 days followed by 51–100 days (16%), 201–300 days (15%), and over 300 days (14%); for company B, the numbers of days are mostly under 10 days (58%) followed by 11–20 days (27%).
• For the number of days from planning to effective implementation of a change request, for company A, the extremes for the number of days between planning and effective implementation are equal at 36%, respectively, under 10 days and above 300 days, followed by the interval of 101–200 days (20%); for company B, the numbers of days between planning and implementation are in the intervals 11–20 days (51%) followed by 21–30 days (35%).

This structure practically confirms the polarization of these variables between the researched companies.

To test if the visible differences between the two companies are statistically significant, a Student’s t-test for independent samples was applied. The results, presented in

Table 7. Results of the Student’s t-test for independent samples.

		Levene’s Test for Equality of Variances		t-Test for Equality of Means
		F	Sig.	t	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	95% Confidence Interval of the Difference
									Lower	Upper
The number of days from creation to planning of change requests	Equal variances assumed	214.384	0.000	35.875	946	0.000	3.614	0.101	3.416	3.812
	Equal variances not assumed			25.367	245.209	0.000	3.614	0.142	3.333	3.894
The number of days from planning to effective implementation of change requests	Equal variances assumed	1099.907	0.000	10.098	947	0.000	1.075	0.106	0.866	1.284
	Equal variances not assumed			6.437	230.443	0.000	1.075	0.167	0.746	1.404

, statistically confirm that the means of the variables are statistically significantly different. Practically, Company A has long-term intervals from change creation to planning and from planning to effective implementation of change requests, and company B has the opposite, i.e., short-term intervals from change creation to planning and from planning to effective implementation of change requests.

By applying the Mann—Whitney test for independent samples, we find that between the two companies, statistically significant differences exist only for the month of effective implementation of change requests (

Table 8. Results of the Mann–Whitney test.

	Null Hypothesis	Test	Sig.	Decision
1	The distribution of the month of proposal/initiation of request for change is the same across categories Company A/Company B	Independent Samples Mann–Whitney U Test	0.873	Retain the null hypothesis
2	The distribution of planned month for implementing of request for change is the same across categories Company A/Company B	Independent Samples Mann–Whitney U Test	0.975	Retain the null hypothesis
3	The distribution of effective implementation month of proposal/initiation of request for change is the same across categories Company A/Company B	Independent Samples Mann–Whitney U Test	0.000	Reject the null hypothesis
4	The distribution of status of change request is the same across categories Company A/Company B	Independent Samples Mann–Whitney U Test	0.000	Reject the null hypothesis

and Figure 9).

Table 8 also presents the results of the Mann—Whitney test for independent samples for a comparison of the status of change requests between the two companies; the results indicate that there are statistically significant differences between the researched companies, and they practically confirm the visible differences shown in Figure 5 and Figure 6.

Regarding customer involvement in change requests, for company A, there is clients involvement in only 25% of the change requests, and the results of a chi-square test indicate that there are no differences according to customer involvement in change requests with respect to the status of change requests (p-value = 0.213). However, the ANOVA test results indicate that, according to customer involvement in change requests, there are statistically significant differences with respect to the number of days from creation to planning of change requests (

Table 9. Results of the ANOVA test by comparing company A, according to customer involvement in change requests.

		Sum of Squares	df	Mean Square	F	Sig.
The number of days from creation to planning of change request	Between groups	112.283	1	112.283	31.493	0.000
	Within groups	759.410	213	3.565
	Total	871.693	214
The number of days from planning to effective implementation of change request	Between groups	7.720	1	7.720	1.338	0.249
	Within groups	1229.210	213	5.771
	Total	1236.930	214

).

Therefore, for Company A, customer involvement in change requests, even a small percentage, does not influence the change request status but it can influence the number of days between creation/proposal of change requests and planning of change requests for implementation.

To determine the best predictors of change request status (as the dependent variable) for each company, we applied multilinear regression models with Enter method and the following independent variables: the number of days from planning to effective implementation of change requests, the number of days from creation to planning of change requests, the effective implementation month of change requests, the month of proposal/initiation of change requests, the planned month for implementing change requests, and the type of process proposed for change. The models are both statistically significant (

Table 10. Results of the ANOVA test for the regression models.

ANOVA a
Company	Model		Sum of Squares	df	Mean Square	F	Sig.
Company A	1	Regression	11,119.494	6	1853.249	19.031	0.000 b
		Residual	20,254.878	208	97.379
		Total	31,374.372	214
Company B	2	Regression	1084.293	6	180.715	6.496	0.000 c
		Residual	20,114.919	723	27.821
		Total	21,199.212	729

^a Dependent variable: Change request status. ^b,c Predictors: (Constant), number of days from planning to effective implementation of change requests, number of days from creation to planning of change requests, effective implementation month of change requests, the month of proposal/initiation of change requests, the planned month for implementation of change requests, type of process proposed for change.

) but explain a small percentage of the total variance of the dependent variable (<30%).

According to the results of the regression models for each company (

Table 11. Regression coefficients for models.

Model a		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta
Company A	(Constant)	23.640	5.118		4.619	0.000
	The month of proposal/initiation of a request for change	−0.033	0.209	−0.009	−0.157	0.876
	Planned month for implementing a request for change	−0.104	0.232	−0.029	−0.448	0.655
	The effective implementation month of a request for change	0.122	0.238	0.032	0.510	0.610
	The type of the process proposed for change	5.880	1.691	0.243	3.476	0.001
	Number of days from creation to planning of a change request	−0.091	0.420	−0.015	−0.217	0.829
	Number of days from planning to effective implementation of change request	2.766	0.304	0.549	9.105	0.000
Company B	(Constant)	39.895	1.065		37.458	0.000
	The month of proposal/initiation of a request for change	0.073	0.163	0.048	0.447	0.655
	Planned month for implementing a request for change	−0.171	0.152	−0.113	−1.126	0.260
	The effective implementation month of a request for change	0.108	0.138	0.071	0.781	0.435
	The type of the process proposed for change	1.139	0.416	0.100	2.741	0.006
	Number of days from creation to planning of change request	−0.940	0.269	−0.173	−3.489	0.001
	Number of days from planning to effective implementation of a change request	−0.142	0.312	−0.023	−0.455	0.649

^a Dependent Variable: Status of change requests.

), the standard coefficient indicates a common predictor for both companies, the type of process proposed for change (p < 0.05) but also different predictors as shown: (a) for Company A, the number of days from planning to effective implementation of change requests (p = 0.000) and (b) for Company B, the number of days from creation to planning of change requests (p = 0.001).

Based on the results from Table 10, respectively, the values of the standardized beta coefficients and the number of days between the three phases of a change request are the most important predictors followed by the type of process proposed for change. By interpreting the values of unstandardized regression coefficients from Table 11, we can conclude the following:

• For company A: (1) Increasing the type of process proposed for change by one unit (from 1 (engineering change) to 2 (corrections)), the status of the change request increases 5.880 units, practically approaching the final status, “closed”; (2) increasing the number of days from planning to effective implementation of a change request by one unit (practically the interval of these days, from <10 days to more than 300 days), the status of the change request increases 2.776 units, practically approaching the final status, “closed”;
• For company B: (1) Increasing the type of process proposed for change by one unit (from 1 (engineering change) to 2 (corrections)), the status of change request increases 1.139 units, practically approaching the final status, “closed”; (2) increasing the number of days from planning to effective implementation of a change request by one unit (practically the interval of these days, from <10 days to more than 300 days), the change request status decreases 0.940 units.

4. Discussion

4.1. Process and Product Change Management

The results of this study demonstrate the importance of the fact that, in an organization in general and in the automotive industry in particular, as a company develops (turnover increases, etc.) and its products are demanded in the market, the number of change requests may increase, and therefore an internal coordination and prioritization system is imperative.

In the present research, such an adaptation was also pursued, namely the implementation of an efficient system for tracking change requests for Company B, starting from an individual analysis (through statistical methods) of this process in Company A, and later, a comparative statistical analysis between the two companies, with the aim of identifying the statistically significant differences that must be taken into consideration when implementing the system in Company B. The basic concept started from the idea that it is not realistic to eliminate engineering changes, and that it is much more important to effectively manage change processes through which the time and cost of implementing and developing the change processes can be reduced; time and cost were identified in this research as specific performance indicators for change requests in the automotive industry. In this research, by highlighting the importance of time and cost for change requests, we practically deny, through our results, the pre-concept often circulated in the specialized literature that engineering changes are predominately perceived to be a problem and not an opportunity [26,27].

Although both companies included in the applied research have the same object of activity, the history, economic performance indicators, and previous results regarding the change processes in the two companies highlight visible differences between Companies A and B.

The need for different strategies for the companies in this study (even if they both operate in the automotive industry and have a similar object of activity) derives from the fact that, for example, the production of electrical parts (the object of activity of Company B) involves much more attention and discipline because the components are present in a wide range and any error leads to a direct increase in costs, an aspect highlighted by the evolution of the economic performance indicators in the two companies for the period 2017–2021; the negative evolution of Company B is evident both through the descriptive statistics values presented in Table 3 as well as the results of the ANOVA test presented in Table 4.

It is also important to create a clear, adapted model of management and organization of change requests in the two companies, which is suitable and aligned with the expectations of each company. In the present research, we identified this by applying, separately for Company A and Company B, multiple linear regression models to identify the best predictor of change request status. Thus, we highlight that the two companies have a common predictor, i.e., the type of the proposed change; however, for Company A (used as a model for implementing the change tracking model for Company B), the most important predictor is the number of days from planning to effective implementation of change requests, while for company B, the best predictor is actually the number of days from creation to planning of change requests.

Practically, the higher the key performance indicator (KPI) specific to the change process in company B increases, the further the change request status moves away from finality. Through these two different predictors for the two study companies, we confirmed, once again, other results of this study; the Student’s t-test results (Table 7) indicated statistical significance for these two performance indicators specific to the change processes in the two companies [28,29].

In addition, the Kruskal–Wallis test results (Table 8 and Figure 9) indicated statistically significant differences between the two companies regarding the effective implementation month of request changes and the change request status; the results confirmed, through other methods of statistical inference, the results above and the significant differences between the two companies. If we drill down and take Company A as a benchmark for both economic and change process-specific performances, the most effective change request KPI is the time between the change initiation date and the change request planning date. According to this KPI, Company B needs to streamline its change process, and this aspect needs to be detailed to the employees and those responsible for the change request in Company B. Most often, models adopted by companies take into account, restrictively in our opinion, the cost dimension in the production strategy, more specifically, production and distribution costs [30] costs that can be reduced by increasing the use of machinery capacity [31] which is recognized in the literature as a cost-based strategy [32] but is the most expensive and difficult to control due to the human factor involved, i.e., the employees who implement, track, and complete the change request.

Key performance indicators (KPIs/KPRs) play a crucial role in a company operating in the automotive industry as they facilitate the accumulation of knowledge and aid in identifying the optimal approach to achieving the organization’s goals. Numerous researchers have proposed various methods for determining KPIs, including manual, semiautomatic, and automatic approaches, which can be applied across different domains. Certain studies have delved into the organization process, the selection criteria for KPIs/KPRs, and even offered a practical example of measuring KPIs/KPRs within the production department. These studies have provided valuable insights into how individuals navigate and analyze complex situations, allowing them to formulate effective strategies and respond accordingly [33].

The need to implement a change process tracking system in Company B also derives from the fact that the annual number of change requests implemented in order to optimize the process/product is approximately three times higher than that in Company A; this significant difference also comes from the field of activity, i.e., Company A operates in the automotive industry through the production of mechanical parts and company B produces electrical parts. Thus, the list of materials that are constantly changing through change requests is much larger for Company B than for Company A. A larger number of change requests implies a longer working time or a greater number of employees who deal with the realization of implementation activities, the determining factors of the complexity of the change request tracking system being both external and internal [34].

Customer involvement is important for providing information for the development of products and/or change processes in industrial markets and are practically beneficial for Company A, as we demonstrated in this paper, the process of effectively monitoring the flow of the change process includes the involvement of customers in requests for change [35].

The results of the ANOVA test indicated statistically significant differences, depending on whether or not customers were involved in the change process, for the number of days from creation to planning of change requests; this specific performance indicator for the change process, for company B, is the better predictor of change request status.

4.2. An Innovative Solution for Change Management as a Predictor of Company Performance: The PCI^® Method

We can declare that the most feasible model for the two companies in this research, and especially for Company B whose economic performance is inferior to that of Company A, is the people-centered change implementation model (PCI^®). The PCI^® method is a comprehensive process of empowering change managers to implement innovations in organizations in a personalized and effective way [3].

The PCI^® method focuses on topics within significant organizational change projects or initiatives and provides a systematic approach to planning, executing, and controlling all particular aspects of a change project. We opt for and recommend this model because at the heart of the PCI^® method is the development of skills, tools, and management processes in planning and implementing change. The process involves the following six main components of the model:

• Formulation of the “shared change purpose” of the current situation, i.e., expected results following the transformation, followed by communication within the organization, so that the employees involved are convinced of the need for change;
• “Effective change leadership”, i.e., creating a network of competent and responsible change leaders;
• Implementation of a “powerful engagement process”, i.e., communication, reward, and development activities to increase the involvement of all employees;
• Committed local sponsors, i.e., middle and line managers who have the skills and motivation to act as role models for their employees;
• Strong personal connection, i.e., models that implement concrete strategies to increase engagement and to develop the skills of their own employees;
• Sustained personal performance, i.e., middle and line managers who are empowered to help their employees implement change so that they fully and effectively embrace it.

4.3. Change Management and Company Performance

Engineering change highlights the role that management plays in the overall change process. Engineering changes are essential and inevitable, and the role of change demand is to ensure that the potential for change is used positively. Change management is based on the timely analysis of changes, which leads to a more efficient way of working and lower costs, and depending on the degree of fulfillment of the requirements, the degree of product quality will improve.

Recent research in the field of change management also indicates the gap between change management and applied practice, as well as the importance of a credible need for change, a plan, and a clear objective doubled by the understanding of this aspect in the organization [36].

Although, in this research, the two companies are competitors, they still have issues in common; among them, the most important issues include satisfying customer demand, promoting innovations, efficient resource allocation, and cost reduction. If change is not handled in an adequate way, there is a possibility that the change process will not achieve the expected objectives [37], and the change request process should be carried out in a transversal way [38].

The internal change process for each of the two companies focuses on quality rather than the speed of change. The reason is that the product is identified as critical to the functionality of a production system, and a system failure can lead to escalations or customer complaints. The management of process/product change offers the possibility of generating specific change KPIs that help the internal tracking of change requests and, where appropriate, the description of improvement measures [39].

Regarding the three essential dimensions of change management (employees, structure, and culture), we make the following recommendations:

• The automotive sector must restructure its research and development organization. While R&D departments currently follow a fairly strict approach, a more streamlined approach may be the answer to the high frequency of innovations expected by customers and regulators.
• Since future innovations will be largely driven by the software that controls the car and the interactions with the passengers, iterative and incremental development and release cycles can also enable this customer experience and product improvements can be made with very high frequency, giving the customer multiple chances to “re-experience” again and again. This can also lead to greater customer loyalty as the customer can install personal “app-like” solutions.
• New business models can help to support regulatory requirements for climate and environmental protection [40,41,42,43].

4.4. Limits, Advantages, and Disadvantages of Process and Product Change Management in the Automotive Industry

Among the limitations of this research we can mention, for Company B, the lack of the possibility of an evaluation of the present economic performances, after the implementation of a structured process of tracking and implementing KPIs specific to the change process according to the model of Company A. Practically, this analysis can be done in the following period, in the medium and long term, to evaluate the concrete results of this process. In the short term, Company B has seen reduced deployment times and much more optimal maintenance and upkeep actions, and errors are discovered much faster, tracked, and resolved consistently.

Regarding the advantages and disadvantages of the proposed process and product change management compared to conventional techniques, we structured them as follows:

• Advantages: (a) An optimized process reduces production costs, increases productivity, and decreases resource wastage; (b) errors and defects are eliminated, leading to reliable and high-quality products; (c) the process contributes to the development of an advanced safety system and improvement of product performance; (d) the process stimulates research and development of new and innovative technologies; (e) by adopting an optimized process the company increases the degree of competitiveness.
• Disadvantages: (a) initiating and implementing a new process may require significant investment in technologies, specialized human resources and advanced equipment; (b) the process can be complex and take a longer period of time to implement; (c) some employees may be reluctant to change or have difficulty adapting to new methods; (d) to successfully manage the process, training and specialization of employees may be required; (e) errors can occur during implementation that can have a huge impact on the final product.

For mass production, it is very important to have an effective change request tracking process. The information processed in the change request process is largely the same in both companies, but there are activities which are particular to each company, i.e., differentiates one company from the other company. The fact that change management is dependent on company factors practically determines/enforces the standardization of the process according to the internal requirements that best suit the change request process. Company B, for example, may choose to customize its system, which is associated with certain cost, but which will give increased efficiency for an optimal process. This research was conducted using solid databases, but the results found do not exhaust the possibility of identifying new factors and continuing, through future research, to collect, for example, the main factors that contributed to success or failure from the perspective of the parties directly involved in the change process [42,43].

5. Conclusions

The results of this study demonstrate the significance of the relationship between organizational growth, especially in the automotive industry, and an increase in change requests. An internal coordination and prioritization system becomes imperative under such circumstances.

A high-performance and tailored system for tracking process and/or product change requests is crucial for rapid assimilation and execution by all involved employees. This system must be adaptable to meet specific needs and requirements. In this research, we implemented an efficient change request tracking system for Company B, based on a statistical analysis of the process in Company A, followed by a comparative analysis between the two companies. The goal was to identify statistically significant differences that should be considered while implementing the system in Company B.

This research emphasizes the importance of effective management of change processes rather than attempting to eliminate engineering changes. Time and cost are identified as specific performance indicators for change requests in the automotive industry, contradicting the perception of engineering changes as predominantly problematic.

While both companies have similar objectives and operate in the automotive industry, they exhibit visible differences in their history, i.e., economic performance and previous results related to change processes. Process and/or product change requests directly influence economic performance indicators and necessitate well-organized production strategies, especially in design and planning stages.

Since Company B deals with electrical parts production, it requires more attention and discipline due to the wide range of components. Any error in this context leads to direct cost increases, as reflected in the economic performance indicators for the period 2017–2021.

This study stresses the need for a clear and tailored model for managing change requests in each company, respecting their organizational cultures, principles, and values. Many models used by companies consider the cost dimension in production strategies, but they often neglect the human factor involved, which can be the most expensive and difficult to control. Key performance indicators (KPIs/KPRs) play a crucial role in the automotive industry for achieving organizational goals, and various methods exist for determining these indicators.

The need for a change process tracking system in Company B arises from the higher number of change requests compared to Company A, primarily due to its field of activity and the constant changes in the list of materials for electrical parts production.

The involvement of customers in change requests is beneficial for Company A, as evidenced by the results of the ANOVA test, which indicated statistically significant differences in the number of days from creation to planning of change requests when customers were involved.

Based on the research results, the most suitable model for both companies, especially Company B with inferior economic performance, is the people-centered change implementation model (PCI^®). This comprehensive process empowers change managers to effectively implement innovations in organizations.

This research offers recommendations related to restructuring the research and development organization in the automotive sector, adopting iterative and incremental development cycles for software-driven innovations, and exploring new business models to meet regulatory requirements.

Despite the solid databases used, the research has limitations, and further studies can explore new factors that contribute to the success or failure in change processes from the perspectives of the parties directly involved.