A Multi-Criteria Assessment Model for Cooperative Technology Transfer Projects from Universities to Industries

Abstract

Cooperative Technology Transfer (CTT) is a technology transfer model where universities and enterprises jointly participate throughout the entire process of technology transfer activities. Most discussions focus on its mechanisms and influencing factors, yet a framework to guide CTT projects in practice is still lacking. This study proposes an assessment model based on the life-cycle of CTT projects, covering the initial cooperation relationship, project management during the mid-term, and technological achievements at the end. The model was evaluated by 14 experts first and then validated through two CTT projects in China. Gray Relation Analysis was employed to calculate the weights of different factors based on their relative importance, while the Dempster–Shafer theory was utilized to combine evidence from various sources and address the uncertainty in the assessment. The results of the case analysis indicate that the attitudes of universities and enterprises are considered critical in influencing the success of CTT projects, while management issues that arise during the projects can pose potential risks. This research serves as an applied exploration and has three functions. Firstly, the model can be used as a feasibility study before the project commences. Secondly, it can be utilized to analyze and improve potential issues during the project. Finally, it can be used for a post-project experience summary.

1. Introduction

Universities are regarded as research centers that aim to cooperate with industries to create innovative products [1,2]. This kind of collaboration can help bridge the gap between industry and academia [3], facilitate the commercialization of knowledge within enterprises [4], and promote socio-economic development in the region [5]. However, research on university–industry technology transfer has largely remained theoretical, discussing various factors influencing technology transfer [6,7,8] but lacking practical application models. Despite the U.S. government’s introduction of Cooperative Research and Development Agreements (CRADAs) to regulate collaborative projects, they primarily serve as agreements and face procedural issues [9].

For the technology transfer project measurement, the Cloverleaf Model proposed by Heslop et al. [10] is an applicable model that focuses on assessing technology maturity and determining which technologies may succeed in the commercialization process. It is evident that evaluating technology maturity requires the potential transferable technology to have reached a particular stage, indicating that the entire project belongs to the technology-driven type. However, this model lacks measures for establishing cooperation relationships and project management between universities and industries for market-driven technology transfer projects.

To address this gap, this paper first summarizes three types of university–industry collaboration models based on the stages at which universities and enterprises participate in technology transfer projects. Then, Cooperative Technology Transfer (CTT) projects, in which universities and enterprises are involved from the outset, collaborate on research and development, and ultimately achieve technology commercialization, are defined. By considering the three main life-cycles of technology transfer projects, this study supplements an assessment model suitable for market-driven CTT projects. In the early stages of the project, the macro-environment, the attitudes and capabilities of the parties involved in the technology transfer, and the technological prospects are evaluated to establish the initial cooperation relationship in the technology transfer project. In the medium stage, the capabilities of the technology transfer team, R&D strategy, contracts, and four categories of management indicators are measured to monitor the project’s progress. In the later stage, the feasibility of the business model, availability and preparation of resources, and achievement transformation team are assessed to determine the success of the technological achievements.

Due to the inclusion of multiple factors in this model, traditional multi-criteria decision-making methods such as the Analytic Hierarchy Process and Hierarchical Decision Modeling face challenges in practical implementation. In this study, Gray Relation Analysis was employed to determine the weights for each factor in the model. When analyzing the validation cases for the model, the Dempster–Shafer evidence theory was utilized to mitigate the limitations arising from differing opinions among multiple experts, thereby reducing the uncertainty in expert decision making.

Through case studies of a completed project and one that is ongoing, this study validates the effectiveness of this model. The application of the assessment model to the completed project indicates that, at the beginning of this three-year project, there was no clear timeframe for research and development, and the funding exceeded the expectations as a result of the ongoing pilot line. However, due to the strong willingness to cooperate, the company had sufficient strength to support such a high-consumption continuous project, and ultimately achieved success. The analysis of an ongoing project shows that the project is also performing poorly in the four categories of management indicators in the medium stage, which will undoubtedly affect the continuation of the project and hinder the subsequent commercialization.

As a supplement to the practical application assessment model for technology transfer projects, this study provides a comprehensive framework for measuring the different stages of CTT projects. In the early stage of the project, this model can be used for a feasibility analysis, providing a comprehensive assessment of technology transfer projects for decision-makers at universities and companies, as well as potential investors. In the middle stage, this model can be used for a phase assessment, identifying potential problem factors and providing insights into whether or not to continue the project. After the project, this model can be used for a project experience summarization, comprehensively analyzing the strengths and weaknesses of the project process.

2. Literature Review and Assessment Model Development

2.1. University–Industry Collaboration Models

University–industry technology transfer refers to the process by which particular inventions or intellectual property rights from academic research are licensed or transferred to profit-making entities through user rights and eventually commercialized [11,12]. This process can be subdivided into different stages [13,14,15], but can generally be categorized into experimental, pilot, and industrialization stages. Based on the timing of participation by enterprises and universities in technology transfer activities, three university–industry collaboration models can be defined, as shown in Figure 1.

The first is the direct transfer model, in which universities are responsible for the experimental stage alone and enterprises are responsible for the pilot stage and the industrialization stage (see Figure 2). The primary manifestation of this model is the technology supplier (generally referring to universities or scientific research units) selling scientific and technological achievements to enterprises, in part or in full, through patents or concessions at one time [16]. The university’s research and development is to make new advancements in the technical field, with the driving force mainly being technology. Once the company has identified a product and demand match, it will decide to acquire the technology product through a transaction [17]. Therefore, universities and enterprises are separated in the direct transfer model. Communication between them is carried out through intermediary organizations such as technology markets. Once the technology has been successfully transferred, the university will quickly reap the benefits and withdraw from the technology transfer process. The pattern of scientific and technological achievements is evident from a technology product perspective. However, because laboratory technology is not researched for the market, not every technological product will be applied to the market transfer process. This makes it more suitable for highly mature scientific and technological achievements, complete technical routes, and precise results.

The second is the self-development model, in which the university is the sole participant, engaging in the entire technology transfer process (see Figure 3). The university utilizes its own advantages to create conditions and establish physical industries based on the existing environment and policies, enabling the smooth transformation and realization of scientific and technological achievements [18]. The driving force behind this technology product may be the market pull or technology push. The main entities using this model are universities with high-tech equipment, abundant scientific research strength, and many scientific research funds. They have developed scientific and technological achievements with a high technological content but, due to the significant technological gap and skepticism about the market, it is challenging for enterprises to accept. Therefore, they choose to establish enterprises themselves. This approach solves technical problems in production and operation and promotes continuous improvement in product development.

For the cooperative research model, or the Cooperative Technology Transfer (CTT), the university and the enterprise have direct contact and participate in all stages of the technology transfer, with universities mainly serving as technical consultants in the industrialization stage (see Figure 4). The cooperation begins due to the enterprise’s demand for technology, and its purpose is to enhance the market competitiveness of enterprises using this technology [19]. Therefore, this model is primarily driven by the market pull. Enterprises put forward their own needs and goals before cooperating. Once both sides agree, the university will send researchers to join in. Researchers use their capabilities to increase the success rate of the technology transfer. In this process, corporate talents can not only learn technical knowledge but also provide market dynamics and improve the commercial value of the products. In this model, the government can establish a platform for resource exchange and joint research, acting as a bridge [20].

CTT is the most widely applicable of the three models regarding both object and product. From the operation process and development perspective, CTT has developed quickly and has gradually become the primary model of university–industry technology transfer activities. Compared with the other two models, the advantages of CTT are as follows: (a) The demand for technical products is proposed by the enterprise, which is the market pull. The research and development of technological products can better adapt to the market and promote economic growth. (b) The research and development stage is jointly completed by the enterprise and the university. Universities can receive sufficient financial support for research and development, while companies can improve their technical capabilities through participation. (c) The university will participate as a technical guide in the transformation stage of technological achievements. Due to the highly technical content, the participation of developers plays an essential role in the successful realization of the technology transfer.

2.2. The Assessment Model for the Cooperative Technology Transfer

The United States government has established Cooperative Research and Development Agreements (CRADAs) to promote collaborative research and development projects between government laboratories and the private sector [9]. However, CRADAs primarily serve as agreements for government laboratories rather than university laboratories. Additionally, although university technology transfer involves the process of productization, commercialization, the industrialization of technology, and the realization of market value [21], there is limited research on the commercialization of technology transfer. Most attention is focused on theoretical discussions, such as the factors influencing technology transfer, participants, and mechanisms [22,23,24,25], while applied research, such as project assessment models, is neglected. Despite Heslop et al. [10] proposing the Cloverleaf Model to assess technological maturity and determine which technologies may succeed in commercialization, it is mainly suitable for measuring university–industry technology transfer under the direct transfer model and the self-development model. As a result, there is still a lack of an applicable and practical evaluation model for assessing university–industry Cooperative Technology Transfer projects.

Considering the characteristics of different stages in CTT projects, this study divides the assessment model into the following three stages: the establishment of cooperative relationships, the cooperative research process, and technological achievement transformation. The first stage typically begins with information gathering and screening. Through the examination of the macro-environment, the attitudes of both the university and the enterprise toward collaboration, and an assessment of the technological entities, this early stage reflects the key drivers and feasibility of the CTT project. Therefore, this stage includes the following four factors: the macro-environment, the university environment and orientation, the enterprise’s comprehensive strength and willingness to cooperate, and technical product evaluation. The second stage emphasizes project management. The technology transfer between universities and industries involves a technical process reflecting the research and development of new technologies or products and the management process, which represents the dynamic transfer of responsibilities among managers. The management process is a supportive function for the technical process, focusing on internal management. It aims to control the progress, cost, and quality of the technology transfer, maximizing the economic value and benefits of the technology [26]. Therefore, this stage includes the following six factors: the technology transfer team, the R&D strategy and contract, as well as organization management, quality management, process management, and capital management. Finally, there is consideration of achievement transformation and commercialization. After the product review, it enters the stage of technological achievement transformation. This mainly involves developing suitable business models based on existing technology, followed by the consolidation of all resources for production. This is accompanied by a professional management team that controls the entire commercialization process. Therefore, this stage includes the following three factors: the feasibility of the business model, the availability and preparation of resources, and the achievement transformation team.

Therefore, based on the factors at different stages, this study proposes a three-stage practical assessment model to measure the CTT project, as shown in Figure 5.

2.2.1. Environment

Environmental factors play a crucial role in the CTT project, representing the external conditions that influence the interaction between universities and enterprises [27]. The government’s attitude and policies are considered the primary source of environmental factors. By implementing a series of policies, the government serves as a platform creator, promoter, and strategy maker for innovation [28]. The government’s ability to absorb, improve, and create new technologies under incentives is significant, as it is responsible for providing infrastructure and creating suitable institutional platforms for communication and knowledge diffusion [29]. Schacht [30] noted that the government may contribute personnel, services, facilities, equipment, intellectual property, and other resources under the umbrella of “personnel, services, and property”. Aghion and Tirole [31] emphasized that government support for R&D is a critical aspect of public policy. The ability of these entities to engage in positive collaborative innovation activities and establish a relatively stable regional innovation network significantly influences the success of university technology transfer [32].

2.2.2. University Environment and Orientation

The environment within universities also influences the establishment of cooperative relationships. Thursby et al. [33] surveyed 62 U.S. universities and found that 71% of inventions required inventor cooperation for successful commercialization. The willingness of inventors to participate in technology transfer activities is mainly related to the guidance provided by the university [34]. However, many universities primarily focus on basic research and teaching that is not directly related to licensing activities, resulting in most researchers’ inventions remaining in the conceptual model or laboratory stage due to a lack of funding or experimental bases for further research [35]. In addition, the incentive measures for researchers, the benefit distribution, and the degree of university support for technology transfer all have a positive impact on the efficiency of the technology transfer [36]. The degree of interdisciplinary communication in universities also affects the technology transfer [37].

2.2.3. Enterprises’ Comprehensive Strength and Willingness to Cooperate

In the CTT project, the university is responsible for research, while the company takes charge of management and market orientation. The company’s management of the technology transfer is a crucial factor in ensuring the smooth development of the project. Many inventions are in the early stages of testing when the contract is signed, and a certain amount of R&D is typically agreed upon in the license contract [31]. Researchers need to develop the product further, as the rejection of a deal by the company could potentially terminate the project. However, accepting a deal also comes with risks. Therefore, a favorable operating environment for the company (including industry and commerce, credit, etc.) provides strong support for dealing with research uncertainty.

2.2.4. Technical Product Evaluation

Technical product evaluation is crucial in the early stages of CTT projects, as it determines the level of support and investment from both universities and enterprises [33,38]. Given the significant role of financial considerations, the feasibility of technology transfer, favorable business prospects, low management complexity, and potential benefits to enterprises, schools, and society can increase the likelihood of investment in the research and development of a specific technology [38,39,40]. Additionally, higher invention value leads to higher license prices, enabling universities to earn more and choose to support the project [39]. Higher royalties stimulate researchers to invest more effort, thereby increasing the likelihood of business success [33]. Consequently, the license price of an invention increases with the success rate of technology transfer projects, creating a greater profit margin for enterprises and achieving a win–win situation. However, each technology has its own life-cycle and level of maturity. Technologies with a lower maturity require a greater investment of resources and time and place higher demands on the technology developers’ research capabilities [27]. Therefore, the alignment between the technological maturity and the expertise of scientific researchers is also a key factor.

2.2.5. R&D Strategy and Contract

Once the university and the enterprise pass the technical product evaluation, they establish a formal partnership, setting detailed product targets and signing technology transfer contracts. Technology transfer contracts encompass various elements, including different types of payments such as milestones, annual payments, copyright fees, and more. Milestones are crucial for ensuring inventor cooperation [41] and play a secondary role in risk sharing. The contract also addresses the ownership of intellectual property rights and the distribution of proceeds among the parties. Instructor inventors who prioritize research over development may not participate if there is no royalty or interest [42]. In addition, increasing the earnings of universities and researchers can enhance their efforts, directly impacting the success rate of the technology transfer [38,43,44]. Therefore, a well-structured R&D program and technology transfer contract can significantly influence the project.

2.2.6. Technology Transfer Team

The technology transfer team serves as the core of the entire project, primarily responsible for the research and development of the technology. Comprising personnel from both companies and universities, the team typically sees the universities focusing on research and the enterprises managing operations. Both parties form a project transfer team with a comprehensive organizational structure [45]. The research capabilities of the R&D personnel are crucial, as highlighted by Li [46], who indicated that high researcher involvement is fundamental for a successful technology transfer. Additionally, the R&D team must possess expertise in essential technologies across all aspects, except for the core technologies, to ensure the smooth progress of the pilot stage [47]. Therefore, the R&D team requires individuals with diverse skills, a clear division of labor, and effective coordination.

2.2.7. Organization Management

Successful technology transfer depends not only on intellectual property creators’ abilities but also on the effective organization of the technology transfer process [48,49]. Effective organizational management is the key factor in enhancing the efficiency of the technology transfer [27]. More specifically, staffing, information updates, and the communication efficiency of the technology team play crucial roles in the performance of university technology transfer [50].

2.2.8. Quality Management

The quality of the technology transfer project encompasses the development and maintenance of the technological achievements [51]. It involves assessing whether the technical achievements adhere to national and industrial design standards and norms, as well as whether the performance factor of the technical achievement aligns with the description of its maturity and advancement in the contract. Additionally, implementing scientific management methods is crucial. The three most popular methods are direct observation, statistical analysis, and regular meeting reports. These methods, respectively, involve the direct examination and judgment of the managed objects, the understanding and assessment of the actual situation of the managed objects through data statistics and analysis, and monitoring the real situation of the managed objects through regular or irregular meetings [52]. Quality management of the technology transfer should revolve around the quality target and involve selecting appropriate methods to strictly control key management points [53].

2.2.9. Progress Management

The management objectives of the technology transfer necessitate both parties to complete specified tasks within predetermined timeframes [54]. When predicting the process of the technology transfer, the total time is broken down into individual stages. Some tasks can be carried out simultaneously, while others are interlinked. The project manager should coordinate relevant personnel to develop an overall control plan for the progress targets by the contract study, identifying key time nodes and important milestones. The task leader should establish this system to monitor the project progress [55]. Implementers of each stage of the technology transfer should complete progress execution report forms, and any issues identified in the reports should be promptly addressed and resolved through coordination [53].

2.2.10. Capital Management

The purpose of technology transfer is to realize the economic value of technological achievements. However, this process involves various input costs, including technology research and development, pilot production and improvement, and more. Capital management aims to estimate the cost of each investment activity in the technology transfer process, aggregating to the total cost. In the early stages of technology transfer project management, targeted market research and inquiries are conducted to prepare a comprehensive cost utilization plan report, offering suggestions for potential savings [56]. Given the significant uncertainties involved in the technology transfer process, allocating sufficient contingency costs is crucial. Therefore, establishing a thorough cost management and supervision system is particularly important [57]. The project manager should regularly compare budgeted costs with actual expenditures, analyze deviations, and decide whether emergency funds are required. At the project’s conclusion, a fund utilization summary should be compiled [53].

2.2.11. Feasibility of Business Model

Technology transfer requires the right business model to generate profits or returns on investment. A business model is a framework for conducting business that enables a company to sustain itself and generate specific revenues [58]. Identifying the target market, market size, and pricing are fundamental elements of commercialization. In some cases, products may target users even before the technology is fully developed. This clarity is crucial for adjusting the direction of the technology transfer, with the company primarily responsible for this aspect. By understanding market dynamics, the company can determine a feasible business model [24].

2.2.12. Availability and Preparation of Resources

The availability of various resources such as funds, materials, production sites, and marketing resources is essential for the transformation of technological achievements [59]. It is important to note that standby resources should not be overlooked, as they are linked to risk prediction and control [60].

2.2.13. Achievement Transformation Team

The technology transformation team should possess extensive business experience and foster effective communication among departments and other business parties. It is important for team members to represent various functions, including suppliers, technical consultants, sales, accounting, and more [27]. Collectively, they contribute to building the new organizational structure. Li [46] emphasized the significance of the participation of technical consultants, particularly R&D teams, in the success of the technology achievement transformation stage. Leaders within the team should be adept at resolving conflicts that may arise during the process. The management team bears the responsibility of overseeing the entire commercialization process, leveraging their expertise to accelerate commercialization and jointly facilitate the technology transformation process.

3. Methodology

In this section, an evaluation framework for CTT is constructed, as shown in Figure 6. A comprehensive review of the literature and a discussion with industry and academic professionals allows for the definition of an assessment tool for CTT. Following the collection of subjective ratings from experts through a questionnaire, multi-criteria decision making is incorporated into the analysis. Gray Relation Analysis is employed to calculate the weights of different factors in the model, considering their relative importance. To further validate the application of the model, this study conducts case studies involving a completed technology transfer collaboration project and an ongoing project. In order to combine evidence from different sources, including the opinions of various experts, and to address the uncertainty inherent in fuzzy assessment [61], the Dempster–Shafer evidence method is utilized. This approach allows for a more comprehensive understanding and consideration of multiple perspectives in the evaluation process.

3.1. Assessment Tool Design

This study develops an assessment tool based on the 13 significant factors identified in CTT projects. Each factor in the model, as depicted in Figure 5, is further expanded into multiple items. The specifics of these items can be found in Appendix A, and the corresponding references are provided in

Table 1. Factors and supporting references.

	Factor	Supporting References
A	Environment	[29]
B	University environment and orientation	[33,36,62]
C	Enterprises’ comprehensive strength and willingness to cooperate	[63]
D	Technical product evaluation	[38,39,40]
E	Technology transfer team	[42,43]
F	R&D strategy and contract	[46]
G	Organization management	[64]
H	Quality management	[51,53]
I	Progress management	[53]
J	Capital management	[53]
K	Feasibility of business model	[58]
L	Availability and preparation of resources	[59]
M	Achievement transformation team	[51]

.

For each factor, several questions are asked to measure the importance/performance of the factor. The questionnaire serves two purposes. First, it is used for an expert assessment of the 13 factors and 145 sub-factors associated with these factors. The score range is from 1 to 10, indicating the level of importance from “not important at all” to “very important”. The importance in terms of weights will be determined based on the assessment by the experts. Secondly, the questionnaire is used to assess the case projects in the case study. In this situation, the score range is from 1 to 5, indicating the level of performance from “extremely bad” to “extremely good”.

3.2. Gray Relation Analysis

The analysis of subjective expert evaluations typically employs the Analytic Hierarchy Process [65,66] and Hierarchical Decision Modeling [67]. However, in cases involving multiple factors, the cumbersome pairwise comparison of factors renders practical implementation unfeasible. Therefore, this study utilizes an enhanced Gray Relation Analysis developed by Cui et al. [68] to compute the weights of different factors. The algorithm fully leverages the subjective information of expert experience judgment values and utilizes a simple mathematical model to objectively calculate the factor weights. The detailed calculation steps are shown below.

Let there be $n$ evaluation indicators and $m$ experts providing empirical judgments on the weights of each indicator, thereby forming empirical judgment data columns for the weights of each indicator, which can be represented in matrix form as follows:

$$X=\left[\begin{array}{c}{x}_1\left(1\right) x_1\left(2\right) \dots x_1\left(m\right)\\ x_2\left(1\right) x_2\left(2\right) \dots x_2\left(m\right)\\ \dots \\ x_n\left(1\right) x_n\left(2\right) \dots x_n\left(m\right)\end{array}\right].$$

Choose the maximum weight value from $X$ as the “common” reference weight value, and assign this value to each expert’s reference weight value, thus forming the following reference data column $X_0$:

$$X_0=(x_0\left(1\right), x_0\left(2\right), \dots , x_0\left(m\right)).$$

Determine the distances between each indicator sequence $X_1, X_2, \dots , X_n$ and the reference data column $X_0$ as follows:

$$D_{0i}=\sum _{k=1}^m{(x_0\left(k\right)-x_i\left(k\right))}^2$$

Calculate the weights of each indicator as follows:

$$w_i=\frac{1}{1+D_{0i}}$$

Finally, calculate the normalized weights of each indicator as follows:

$${w_i}^{*}=\frac{w_i}{\sum _{i=1}^n{w}_i}$$

3.3. Dempster–Shafer Evidence

The application of the calibrated CTT project assessment tool to specific technology transfer projects requires obtaining evaluations from multiple experts for the same project to compensate for the potential bias and limitations of individual experts’ experiences and knowledge. This study employs Dempster–Shafer evidence to mitigate the limitations of differing opinions among multiple experts, thereby reducing the uncertainty in expert decision making.

However, since the obtained evidence often contains conflicting information, Dempster’s rule of combination is often invalid [69]. So far, the modification schemes for Dempster–Shafer theory defects have mainly been divided into two categories. The first is the modification of combination rules and the second is the preprocessing of evidence sources. It is believed that the second method is more suitable because modifying the combination rule will destroy the original method’s associative law and commutative law [70]. For the treatment of evidence sources, the most typical method is Murphy’s method [71] of the equal distribution of evidence sources, which has been shown to greatly increase the accuracy of evidence fusion [72]. Therefore, this study evaluates the case by using Murphy’s improved Dempster–Shafer evidence theory. The detailed calculation steps are shown below.

The frame of discernment is as follows:

$$\mathsf{\Theta }=\left\{h_k|k=1, 2, 3, 4, 5\right\}=\left\{extremly bad, bad, neutral, good extremly good\right\}$$

where each variable represents one of the five levels of the performance evaluation factors. The basic probability assignment is utilized to articulate individuals’ confidence in the evidence. In this study, Gray Relational Analysis theory is employed to establish the basic probability assignment, satisfying the following conditions:

$$m\left(\mathrm{\varnothing }\right)=0$$

$$\sum _{\mathrm{A}\subseteq \mathsf{\Theta }}m\left(\mathrm{A}\right)=1$$

The function $m\left(\right)$ is referred to as the mass function, where $m\left(\mathrm{A}\right)$ denotes the degree of support of the evidence for proposition A.

Then, the synthetic rule of Dempster–Shafer theory is used to perform the orthogonal calculation of the basic probability assignment of the evidence. For example, suppose and are the basic probability assignments for identifying evidence E1 (expert 1) and E2 (expert 2) in the frame, then:

$$m(\mathrm{A})=\left\{\begin{array}{c}\frac{1}{1-h}{\displaystyle \sum _{{\mathrm{A}}_i\cap {\mathrm{B}}_j=\mathrm{A}}}{m}_1\left({\mathrm{A}}_i\right)m_2\left({\mathrm{B}}_j\right),\mathrm{A}\ne \mathrm{\varnothing }\\ 0, \mathrm{A}=\varnothing \end{array}\right.$$

$$h=\sum _{{\mathrm{A}}_i\cap {\mathrm{B}}_j=\varnothing }{m}_1\left({\mathrm{A}}_i\right)m_2\left({\mathrm{B}}_j\right)$$

where $h$ represents the degree of conflict between evidence E1 and E2. According to Murphy’s improved method, before using the Dempster–Shafer evidence, it is distributed to the average as follows:

$$m^{\ast }(i)=\sum _{n=1}^h{m}_h\left(i\right)$$

The degree of support for proposition A under the Dempster–Shafer evidence can finally be obtained as follows:

$$\mathrm{D}\mathrm{S}(k_i)=\frac{{m^{\ast }\left(k_i\right)}^2}{\sum _{i=1}^5{m^{\ast }\left(k_i\right)}^2}$$

4. Results

4.1. Factor Weight Allocation

Since CTT involves not just a single industry, this study invited 14 experts with experience in CTT projects in China from various industry fields to ensure the effectiveness and generalizability of the assessment tool. Their basic information is listed in

Table 2. The background of the 14 experts.

Expert	Industry	Years of Experience of Work in China	Years of Experience in CTT Projects in China	Number of CTT Projects in China
1	Electrical engineering	18	7	5
2	Manufacturing	16	6	4
3	Textile	30	16	9
4	Computer	6	3	3
5	Automobile manufacturing	10	5	4
6	Education—marine Technology	35	6	5
7	Education—machinery	30	12	8
8	Computer	14	7	6
9	Education—electrical Information	18	15	7
10	Manufacturing	10	10	7
11	Education—chemistry	7	3	2
12	Education—energy	20	15	15
13	Education—machinery	20	10	5
14	Manufacturing	10	5	5

. Data were collected during separate correspondence with each expert.

The analysis of the data using the GRA method involves two steps. The first step is to calculate the weights of different sub-factors within each factor separately, obtaining the experts’ perceptions of the importance of different sub-factors and providing more detailed insights for practical improvement. The second step involves treating all of the sub-factors as a whole to calculate the proportion of each factor in the overall model, providing a more macro-assessment of the overall technology transfer project.

Due to space limitations, this study provides a detailed calculation process using part G ‘Organization management’ as an example. There are seven sub-factors in this factor. This study takes G1–G7 as a whole and obtains the following scoring matrix:

$$\mathrm{X}={(\mathrm{X}1,\mathrm{X}2,\dots ,\mathrm{X}7)}^T$$

$$\mathrm{X}=\left[\begin{array}{ccc}x1\left(1\right)& \cdots & x1\left(14\right)\\ ⋮& \ddots & ⋮\\ x7\left(1\right)& \cdots & x7\left(14\right)\end{array}\right]$$

The specific data are shown in

Table 3. Experts’ ratings of ‘G: Organization management’.

	E1	E2	E3	E4	E5	E6	E7	E8	E9	E10	E11	E12	E13	E14
G1	8	7	8	8	9	7	9	8	9	10	8	8	8	7
G2	9	8	7	7	7	7	8	7	7	8	7	8	8	8
G3	8	8	8	8	8	7	9	7	7	10	10	8	9	6
G4	8	8	8	7	7	6	8	7	7	9	7	8	7	6
G5	9	9	8	9	8	8	9	7	9	9	10	8	9	8
G6	8	9	8	8	7	7	8	8	8	9	8	8	8	8
G7	7	8	8	7	7	7	7	7	7	8	7	8	8	7

.

A maximum weight value from all X variables is selected as the “common” reference weight value, which is assigned to each expert’s reference weight value to form the reference data column X0, as follows:

$$\begin{gathered} \mathrm{X}0=\left(\mathrm{x}0\left(1\right),\dots ,\mathrm{x}0\left(14\right)\right) \\ =\left(\mathrm{10,10,10,10,10,10,10,10,10,10,10,10,10,10}\right) \end{gathered}$$

Calculate the distance between the sequence X1, X2, …, X7 and the reference data in column X0 as follows:

$$D_{0i}=\sum _{k=1}^{14}{\left(x0\left(k\right)-xi\left(k\right)\right)}^2$$

For example:

$$\begin{array}{ll}{D}_{01}=& {\displaystyle \sum _{k=1}^{14}}{(x0\left(k\right)-x1\left(k\right))}^2={(10-8)}^2+{(10-7)}^2+{(10-8)}^2+{(10-8)}^2+{(10-9)}^2+{(10-7)}^2\\ & +{(10-9)}^2+{(10-8)}^2+{(10-9)}^2+{(10-10)}^2+{(10-8)}^2+{(10-8)}^2+{(10-8)}^2\\ & +{(10-7)}^2=58\end{array}$$

The same algorithm can be used to obtain the following:

$$D_{02}=\sum _{k=1}^{14}{(x0\left(k\right)-x2\left(k\right))}^2=88$$

$$D_{03}=\sum _{k=1}^{14}{(x0\left(k\right)-x3\left(k\right))}^2=69$$

$$D_{04}=\sum _{k=1}^{14}{(x0\left(k\right)-x4\left(k\right))}^2=107$$

$$D_{05}=\sum _{k=1}^{14}{(x0\left(k\right)-x5\left(k\right))}^2=36$$

$$D_{06}=\sum _{k=1}^{14}{(x0\left(k\right)-x6\left(k\right))}^2=60$$

$$D_{07}=\sum _{k=1}^{14}{(x0\left(k\right)-x7\left(k\right))}^2=101$$

Calculate the weight of each sub-factor as follows:

$$wi=\frac{1}{1+D_{0i}}$$

Then:

$$w1=\frac{1}{1+58}=0.016949153$$

$$w2=\frac{1}{1+88}=0.011235955$$

$$w3=\frac{1}{1+69}=0.014285714$$

$$w4=\frac{1}{1+107}=0.009259259$$

$$w5=\frac{1}{1+36}=0.027027027$$

$$w6=\frac{1}{1+60}=0.016393443$$

$$w7=\frac{1}{1+101}=0.009803922$$

Calculate the normalized weight of each sub-factor as follows:

$${wi}^{\ast }=\frac{wi}{\sum _{i=1}^{7}wi}$$

Then:

$${\sum }_{i=1}^{7}wi=0.104954472$$

$${w1}^{\ast }=\frac{0.016949153}{0.104954472}=0.161490522$$

$${w2}^{\ast }=\frac{0.011235955}{0.104954472}=0.107055515$$

$${w3}^{\ast }=\frac{0.014285714}{0.104954472}=0.13611344$$

$${w4}^{\ast }=\frac{0.009259259}{0.104954472}=0.088221674$$

$${w5}^{\ast }=\frac{0.027027027}{0.104954472}=0.257511914$$

$${w6}^{\ast }=\frac{0.016393443}{0.104954472}=0.156195751$$

$${w7}^{\ast }=\frac{0.009803922}{0.104954472}=0.093411184$$

Then, the weight allocation under ‘G: Organization management’ is obtained, as shown in

Table 4. Weight allocation of ‘G: Organization management’.

	G1	G2	G3	G4	G5	G6	G7
Weight	0.1615	0.1071	0.1361	0.0882	0.2575	0.1562	0.0934

.

This example takes G1–G7 as a whole and calculates the weight of each sub-factor under the G factor. Then, the same method is used to calculate the weights for each factor from A to M, respectively.

At the same time, a total of 145 sub-factors of A1–M11 as a whole is obtained. The calculated weights of each factor are presented in

Table 5. The weights of the 13 factors.

	Factor	Weights
A	Environment	6.04
B	University environment and orientation	10.94
C	Enterprises’ comprehensive strength and willingness to cooperate	8.51
D	Technical product evaluation	9.35
E	Technology transfer team	13.93
F	R&D strategy and contract	11.07
G	Organization management	3.36
H	Quality management	9.45
I	Progress management	4.95
J	Capital management	3.95
K	Feasibility of business model	4.25
L	Availability and preparation of resources	7.24
M	Achievement transformation team	6.97
Sum		100

, and the results for each sub-factor are shown in Appendix A.

4.2. Case Study 1—Composite Spinning Technique

The first case reports on the technology transfer from University B to Company A in China. The project occurred between 2004 and 2007. This section provides a retrospective evaluation of the CTT process for this case. Two core members who participated in the project were invited to give feedback on the case, with their information shown in

Table 6. Core members of Case 1.

	Industry	Company	Position	Project Responsibility
1	Textile	Company A	Production manager	Resource management, team management, and technology communication
2	Education—textiles	University B	Professor	Research and development

.

The research steps are as follows:

(1)

Description of the early stage of the project (establishment of the cooperative relationship).

(2)

Description of the medium stage of the project (cooperative R&D process).

(3)

Description of the later stage of the project (transformation of technical achievements).

(4)

Analysis of the evaluation results of the core members.

4.2.1. Early Stage: Establishment of Cooperative Relationship

Before 2004, Company A faced two problems. The first was a limitation in the spinning operations: 66 pieces of wool could only be spun into 60 yarns, which hindered their efficiency. The second issue was related to the spinning process, which restricted the length, grade, and variety of raw materials. Specifically, cotton ball fibers had to be more than 16 mm long or they would be wasted, leading to inconvenience and resource consumption in production. In response, Company A invited professors from University B to visit its factory and present its technical requirements. The goal was to break through traditional textile technology and improve the production efficiency. It was anticipated that the development of this technology would not only impact the production and operation of Company A but also bring about a breakthrough in the entire textile industry. Therefore, this technical requirement aligns with the market-pull type, which is consistent with the original intention of the cooperative research mode.

Although the project presented significant challenges, University B decided to collaborate with Company A. This decision was not solely because University B possessed a leading research team in the textile industry. University B also actively fostered an environment conducive to technology transfer. Despite the difficulty of realizing the new technology with Company A’s existing technology at that time, Company A demonstrated a spirit of adventure and innovation, as well as the financial resources to support this uncertain project. As a result, after communication between the enterprises and the university, they deemed this technical product acceptable for several reasons: (1) it had comparative advantages compared to other technologies; (2) the more compatible the existing regime, the faster the technology transfer would be; (3) it was not overly complicated to use; (4) it could be tested before use, allowing for the observation of adoption results before the full implementation. Consequently, cooperative research and development commenced to expand the application range of textile raw materials and increase the variety of textiles.

4.2.2. Medium Stage: Cooperative R&D Process

Following the establishment of the collaboration, a technology transfer team was formed. After careful consideration, the researchers from University B proposed their theoretical model, which involved using a short-flow embedded composite spinning technology. The technology was still experimental and in its early stages of development. Company A needed to modify the existing technical scheme and conduct trial production to validate the technology’s rationality and feasibility. Despite facing significant risk, Company A spared no effort in mobilizing the best equipment and the most exceptional technical personnel within the enterprise to repeatedly adjust the technological process based on the theoretical design.

Additionally, several project managers from Company A were assigned to provide personal guidance and integrate the technical expertise of each link in the production workshop to create samples. After three years of research and development, everything from spinning to tailoring was deemed a success. In comparison to the original process, this technique achieved the following breakthroughs: (1) overcoming the limit of the yarn count in the original spinning process, increasing it from 60 yarns to 80 yarns; (2) addressing the shortage of textile raw materials and maximizing the resource utilization. This allowed for the combined production of cotton, wool, rabbit down, and previously non-spinnable down fiber to form a higher yarn count on the spinning machine.

4.2.3. Later Stage: The Transformation of Technical Achievements

The commercialization of scientific and technological achievements is a highly complex process. Company A commenced a comprehensive transformation of its production scheme after successfully testing the technology. This stage primarily involved the implementation and operation of the technology within the company. However, the involvement of the R&D team was also crucial. During the large-scale application of the technology, University B assigned scientific researchers to provide technical guidance to the technical staff of the enterprise on multiple occasions and continued to monitor the progress until full implementation was achieved.

4.2.4. Evaluation Results and Analysis

Due to space limitations, this study provides a detailed calculation process using part G ‘Organization management’ as an example. There are seven sub-factors in this factor. The first core member’s score is {a₁, …, a₇}. Suppose the score at grade U_j is {a_j1, …, a_jm}. Then, the basic probability allocation function of trust level U_j is as follows:

$$p_j=\frac{{\sum }_{m=1}^k{w}_{jm}\times j}{{\sum }_{i=1}^7{w}_i{a}_i}=\frac{w_{j1}\times j+w_{j2}\times j+\dots +w_{jk}\times j}{w_1{a}_1+w_2{a}_2+\dots +w_7{a}_7}$$

where $a_i$ represents the score of each sub-factor from G1 to G7 and $w_i$ represents the weight of each sub-factor from G1 to G7.

Let us say that there are k sub-factors in the score of j. So, $w_{jm}$ is equal to the weights of the k sub-factors one-by-one. We use the first expert score from Case 1 as an example. The scores and the weights of the G factors are calculated in

Table 7. Evaluations of Case 1 from expert 1 for ‘G: Organization management’.

	Score of Each Sub-Factor	Weight of Each Sub-Factor
G1	4	0.1615
G2	4	0.1071
G3	5	0.1361
G4	3	0.0882
G5	4	0.2575
G6	4	0.1562
G7	3	0.0934

.

Then, the basic probability allocation function of the trust score i is represented by m_j(Ui), where j represents expert j. Then, the assignment functions of expert 1 are as follows:

$$m_1\left(\mathrm{U}1\right)=0$$

$$m_1\left(\mathrm{U}2\right)=0$$

$$\begin{array}{ll}{m}_1\left(\mathrm{U}3\right)=& \frac{0.0882 \times 3 + 0.0934 \times 3}{0.1615 \times 4 + 0.1071 \times 4 + 0.1361 \times 5 + 0.0882 \times 3 + 0.2575 \times 4 + 0.1562 \times 4 + 0.0934 \times 3}\\ & =0.137792705\end{array}$$

$$\begin{array}{ll}{m}_1\left(\mathrm{U}4\right)=& \frac{0.1615 \times 4 + 0.1071 \times 4 + 0.2575 \times 4 + 0.1562 \times 4}{0.1615 \times 4 + 0.1071 \times 4 + 0.1361 \times 5 + 0.0882 \times 3 + 0.2575 \times 4 + 0.1562 \times 4 + 0.0934 \times 3}\\ & =0.690107019\end{array}$$

$$\begin{array}{ll}{m}_1\left(\mathrm{U}5\right)=& \frac{0.1361 \times 5}{0.1615 \times 4 + 0.1071 \times 4 + 0.1361 \times 5 + 0.0882 \times 3 + 0.2575 \times 4 + 0.1562 \times 4 + 0.0934 \times 3}\\ & =0.172100276\end{array}$$

Then, expert 2’s basic probability distribution function is obtained, as shown in

Table 8. The basic probability assignment function.

G	m2(U1)	m2(U2)	m2(U3)	m2(U4)	m2(U5)
	0	0	0	0.390668	0.609332

.

From this, the basic probability distribution function of each of the 13 factors and the whole model of each expert can be calculated. Let us denote the basic probability distribution of experts j for Ui as m_j(Ui).

Then, Dempster–Shafer theory is used to combine the evidence from various sources and address the uncertainty. Case 1 was evaluated by two experts, so the source of the evidence was processed according to Murphy’s improved method, taking the average of m(Ui) as m*(Ui) as follows:

$$m^{\ast }\left({\mathrm{U}}_1\right)=\frac{m_1\left({\mathrm{U}}_1\right)+m_2\left({\mathrm{U}}_1\right)}{2}$$

So, [m*(U1), m*(U2), m*(U3), m*(U4), m*(U5)] can be calculated as follows:

$$m^{\ast }\left(\mathrm{U}1\right)=0$$

$$m^{\ast }\left(\mathrm{U}2\right)=0$$

$$m^{\ast }\left(\mathrm{U}3\right)=0.068896352$$

$$m^{\ast }\left(\mathrm{U}4\right)=0.540387343$$

$$m^{\ast }\left(\mathrm{U}5\right)=0.390716304$$

According to the fusion law of Dempster–Shafer theory:

$$\mathrm{D}\mathrm{S}\left({\mathrm{U}}_{\mathrm{i}}\right)=\frac{{m^{\ast }\left({\mathrm{U}}_{\mathrm{i}}\right)}^2}{{m^{\ast }\left({\mathrm{U}}_1\right)}^2+{m^{\ast }\left({\mathrm{U}}_2\right)}^2+{m^{\ast }\left({\mathrm{U}}_3\right)}^2+{m^{\ast }\left({\mathrm{U}}_4\right)}^2+{m^{\ast }\left({\mathrm{U}}_5\right)}^2}$$

So, the calculated results are as follows:

$$\mathrm{D}\mathrm{S}\left(\mathrm{U}1\right)=0$$

$$\mathrm{D}\mathrm{S}\left(\mathrm{U}2\right)=0$$

$$\mathrm{D}\mathrm{S}\left(\mathrm{U}3\right)=0.010561748$$

$$\mathrm{D}\mathrm{S}\left(\mathrm{U}4\right)=0.649761047$$

$$\mathrm{D}\mathrm{S}\left(\mathrm{U}5\right)=0.339677205$$

From this, the basic probability of each result Ui from A to G by the Dempster–Shafer theory of factors is obtained, and then the performance grade of the project according to the maximum membership principle is determined, with the results shown in

Table 9. Evaluation results of Case 1.

		U1	U2	U3	U4	U5
A	Environment	0	0	0.2241	0.7759	0
B	University environment and orientation	0	0	0.3252	0.3113	0.3635
C	Enterprises’ comprehensive strength and willingness to cooperate	0	0	0.0001	0.0117	0.9882
D	Technical product evaluation	0	0	0	0.0137	0.9863
E	Technology transfer team	0	0	0.0002	0.0966	0.9032
F	R&D strategy and contract	0	0	0.0117	0.0055	0.9828
G	Organization management	0	0	0.0106	0.6498	0.3397
H	Quality management	0	0	0.0029	0.0212	0.9760
I	Progress management	0	0.0118	0.9702	0.0180	0
J	Capital management	0	0.5085	0.4637	0.0278	0
K	Feasibility of business model	0	0	0	0.0693	0.9307
L	Availability and preparation of resources	0	0	0.0261	0.7160	0.2579
M	Achievement transformation team	0	0	0.0026	0.1233	0.8741
Whole		0	0.0004	0.0285	0.1140	0.8571

Note: The bold numbers in the table highlight the maximum basic probability and corresponding Ui of the factor.

.

Based on the maximum membership principle, it can be observed that most of the factors behave at the levels of “U4 = good” and “U5 = extremely good”. This indicates that the project’s overall performance is excellent, which is also consistent with the actual situation.

Firstly, the influence of the environment (score in U4 = good) on the technology transfer was subtle. The government’s strong support and the creation of a social environment for technology transfer made such cooperative projects possible. Additionally, the government provided funds and credit preferences during the project, which greatly helped alleviate the enterprise’s pressure.

The influence of University B (score in U5 = extremely good) on the project cannot be ignored. University B not only focuses on teaching but also actively encourages scientific researchers to participate in technology transfer activities. Over the course of three years, professors were allowed to dedicate a significant portion of their working time to research and were provided with the necessary support. Furthermore, the university organized textile technology exchange meetings on multiple occasions to promote the project’s technical progress. As a result, the composite textile technology has significantly enhanced the school’s reputation.

The enterprise’s support (score in U5 = extremely good) played a crucial role in this project. The cost of the three-year project, in terms of capital, workforce, and resources, cannot be underestimated. Additionally, in the early stages of development, technological uncertainty was high. In the face of such a significant risk, Company A demonstrated unwavering determination, which ultimately led to the success. During the pilot stage, despite facing substantial production adjustments, Company A chose to support the testing process. This required strong overall capability and a steadfast commitment to the technology.

The enterprise proposed this technical product (score in U5 = extremely good) in response to market demand. Although the uncertainty was significant, the application prospects were excellent. This is also the reason why Company A chose to collaborate on this project.

The research strength of the technology transfer team (score in U5 = extremely good) was the most important factor in determining the project’s success. In addition to professors from University A, numerous other textile experts from the academic world participated in seminars to discuss the technology transfer plans. Furthermore, the enterprise not only assigned a large number of technical staff with mature processes to participate in the project but also assigned a substantial number of outstanding talents in finance, machinery, and other relevant fields to participate. It can be said that this was a team with a strong overall capability.

A reasonable R&D strategy (score in U5 = extremely good) was essential for the technology transfer. The incentive clauses in the contract also promoted the enthusiasm of the R&D personnel. The organizational cohesiveness (score in U4 = good) of the R&D team was high, and the researchers’ assumptions about technology were effectively communicated to the companies. The company and the university maintained efficient communication. Quality management (score in U5 = extremely good) was essential to the R&D process. After proposing the technical hypothesis, the company and the university jointly conducted the pilot experiment, and the success of the implementation process represents a significant milestone in R&D.

During the stage of technology achievement transformation, the business model (score in U5 = extremely good) of this technology was determined from the outset, aiming to improve the production efficiency for self-marketing technology products. Therefore, the preparation of resources (score in U4 = good) and the organization of the technology transfer team (score in U5 = extremely good) were determined during the scientific research stage. Coupled with the careful guidance of scientific researchers, the transformation results were ideal.

However, the results of ‘Process management’ and ‘Capital management’ are not ideal (to the degree of U3 = neutral and U2 = bad). This is also consistent with the reality. The project faced significant challenges, primarily attributed to technological limitations. On one hand, the existing technology of Company A made it difficult to implement the new technology. On the other hand, the technology proposed by University B was still in the experimental stage and in the early stages of development. Therefore, Company A needed to modify the existing technological solution and conduct pilot production to validate the feasibility and viability of the technology. Additionally, at the beginning of the project, there was no clear timeframe for research and development. This resulted in the ongoing pilot phase of this three-year project exceeding the anticipated time and budget constraints. However, due to the firm’s strong cooperation intention, the enterprise had enough strength to support such a high-consumption continuous project, leading to the ultimate success. Also, in some respects, the management of funds and schedules was insufficient, mainly due to the high level of technical uncertainty.

4.3. Case Study 2—Intelligent Production Control System

In most situations, the decision-makers and evaluators of CTT projects often prefer to conduct a comprehensive assessment in the early stages of a project. Therefore, Case Study 2 presents an ongoing project. The partners involved include Company C and University D, and the project commenced in November 2019 in China. In April 2020, we conducted an investigation and evaluation using the model presented in this study. The project team made some improvements to the evaluation results to determine if the model was instructive. All of the analyses were based on the project team’s extensive experience in CTT and their ability to assess potential future outcomes based on the current situation. The research process is as follows:

(1)

Project background and technical requirements.

(2)

Evaluation results and analysis.

(3)

Adjustment and completion.

4.3.1. Project Background and Technical Requirements

In November 2019, Company C proposed a collaboration request to University D to jointly develop a production control system. The goal of their client, Company E, a cement company, was clear. After conducting a joint investigation at Company E, they identified the following issues: the company’s DCS, ERP, OA, and other systems were isolated from each other, the information collection process was cumbersome, there existed multiple sets of standardized software systems reducing the operational efficiency, the production plan was not reasonable, and the efficiency of using manual reports needed improvement. The existing energy management system made it challenging to collect energy consumption data and accurately calculate the energy consumption in real-time, making subsequently refined statistics more challenging. Quality management was time-consuming and it was difficult to track the root cause after a quality incident.

Due to University D and Company C’s similar experience in technology transfer and the relatively high maturity of this technology, they quickly reached an agreement. After the communication, they decided to adopt the design concept of a bottom-up three-level architecture model, limited standardization, and deep data mining at the basic level. As a result, they designed a smart production management and control system covering production management, equipment management, and intelligent decision making. This cooperative transfer process was driven by the market pull, with Company E being their target market. They aim was to improve the production and operation system of Company E to enhance its economic efficiency.

4.3.2. Evaluation Results and Analysis

As of April 2020, the project team had completed 17 modules, including the production plan execution, key equipment management, and power energy statistics. It can be said that the operation was relatively smooth. This study invited four core members (with detailed information shown in

Table 10. Core members of Case 2.

	Industry	Company	Position	Project Responsibility
1	Education—computer	University D	Professor	Head of R&D
2	Education—computer	University D	Ph.D. student	A member of the R&D team
3	Computer	Company C	Manager	R&D and project hub
4	Computer	Company C	Operations manager	Deliver customer needs and team management

) to evaluate their performance in the first half and obtained the following results.

According to Murphy’s improved Dempster–Shafer law, the results are obtained in

Table 11. Evaluation results of Case 2.

		U1	U2	U3	U4	U5
A	Environment	0	0	0	0.2004	0.7996
B	University environment and orientation	0	0	0.0001	0.3212	0.6787
C	Enterprises’ comprehensive strength and willingness to cooperate	0	0	0.0146	0.9410	0.0444
D	Technical product evaluation	0	0	0.0029	0.8752	0.1219
E	Technology transfer team	0	0	0.0057	0.4091	0.5852
F	R&D strategy and contract	0	0	0.0052	0.5272	0.4676
G	Organization management	0	0.7810	0.0789	0.1401	0
H	Quality management	0	0.2815	0.3654	0.3129	0.0402
I	Progress management	0	0.3162	0.6361	0.0477	0
J	Capital management	0	0.3507	0.4009	0.2485	0
K	Feasibility of business model	0	0	0.0143	0.7338	0.2519
L	Availability and preparation of resources	0.0004	0.4136	0.3159	0.2701	0
M	Achievement transformation team	0	0.0504	0.7680	0.1816	0
Whole		1 × 10−6	0.0196	0.0490	0.5354	0.3961

Note: The bold numbers in the table highlight the maximum basic probability and corresponding Ui of the factor.

.

Overall, this project achieved good enterprise, school, and product evaluation results, with an overall score leaning towards U4 = good and U5 = extremely good. This indicates that the project has great potential for successful completion.

However, the management score was not satisfactory, mainly falling within the range of U2 = bad to U3 = neutral. Firstly, ‘Capital Management’ is at the U3 = neutral level. From January to April, project spending exceeded expectations. This was mainly due to the lack of field knowledge of Company E during the design phase, resulting in the server being unable to handle such a large amount of data. The server connection failed, leading to a slow user response. As a result, the project team had to spend more money to upgrade the hardware. As the reserve fund was not expected to be sufficient, there was a fund deviation. However, Company C had a high tolerance for project deviation, so it did not affect their partnership. In a project, money, quality, and schedule are interdependent. Unreasonable hardware design was also related to quality management (score in U3 = neutral). This was due to the initial hardware quality not meeting the requirements. Additionally, quality management also includes the inspection of equipment and systems. Supervision of the project was infrequent, which can also cause problems. For example, at a specified time node, the installation of electricity meters on site was not confirmed, resulting in the failure to complete the task of this node on time. This represents both a quality deviation and a schedule delay. Furthermore, the project team often had to reiterate due to interface problems and some system bugs, resulting in wasted time and money. As a result, the team’s process management is only at the U3 (neutral) level. From an objective standpoint, the current completion of the project is delayed by half a month compared to expectations.

In addition, the most unsatisfactory performance is in ‘Organization management’, which is at the level of U2 = bad. This is mainly due to the fact that most of the funding, schedule, and quality problems were caused by miscommunication and task misallocation within the team. For example, the absence of the electricity meter in the example above was due to ineffective communication between the researcher and the equipment collector. The reiteration of the system was also caused by technical personnel not understanding the customer’s requirements. Due to the remote location, the technology transfer team communicated with the customer, Company E, mainly through video conferences, which made it difficult to fully understand the customer’s needs. Furthermore, field equipment maintenance and testing personnel in the project worked at Company E, while project team members were far away. Their meeting times were not fixed, leading to incomplete information transmission.

5. Discussion and Implications

This study introduces an assessment model for CTT projects aimed at evaluating the performance of each stage of technology transfer. The model serves the following three main functions: (1) before achieving cooperation, parts A to D of this model can be used to assess the investment worthiness; (2) at each stage of the technology transfer, this model can be utilized to evaluate the project performance and gain insights for project modification; (3) upon completion of the cooperation, this model can be used to reflect on the project and summarize the experience. As a result, this study provides the following theoretical and practical implications.

5.1. Theoretical Implications

Firstly, the assessment model highlights the importance of evaluating the content before cooperative research and development. It is not merely an assessment of the market value of technology projects; it encompasses the company’s willingness to cooperate, the economic capacity, and the university’s support for technology transfer as crucial criteria. The success of the technology project in Case 1 was attributed to the company’s decision to provide substantial support in the face of uncertainty, indicating that the attitude and level of support from the enterprises largely determines the smooth progress of a project [46]. These early-stage factors account for 34.84% of the total weight, demonstrating the significance of early project assessment in the views of experts.

Secondly, the medium stage of the technology transfer is the most critical, with the combined importance ratio of the six factors reaching 46.71%. Notably, the technology transfer team and R&D strategy and contract are considered key factors in technology transfer projects, with weights of 13.93% and 11.07%, respectively. Considering that these two factors are in the transition from the early stage to the medium stage, their importance to the overall technology transfer project is undeniable. Additionally, Cooperative Technology Transfer shares common characteristics with all projects. Therefore, this study introduces the concept of project management and defines four management factors. In the two cases presented in this study, despite the lack of appropriate management tools during the R&D process, the overall evaluation of the projects remained satisfactory. However, the combined importance ratio of the four management factors still reached 21.71%, indicating that, despite the enterprise’s strong tolerance, poor project management will inevitably bring potential risks.

Finally, the model considers not only the transfer of technology from universities to enterprises but also to the market. In addition to the traditional business processes, the commercialization of technology should also reflect the unique aspect of technology transfer, namely, the application of high-tech products. According to Jensen and Thursby [42], at least 71% of inventions require the further involvement of researchers for successful commercialization. The case study also highlights the importance of this view. The primary reason why a company needs to collaborate with a school is that its technical capabilities cannot keep pace with the rapid changes in the market. In other words, enterprises sometimes lack the capacity to develop high-tech products independently. For highly complex technologies, enterprises need to fully assimilate scientific and technological knowledge to bring them to market.

5.2. Practical Implications

This model provides a feasible framework for evaluating CTT projects and has yielded the following implications. For initiators or managers of technology transfer projects, this model offers an analysis of the importance of different factors for project success, enabling the project team to adjust the strategy accordingly within limited resources, for example, placing the technology transfer team (13.93%) at a higher priority than organizational management (3.36%). Additionally, it allows for the assessment of problematic factors during the project and the optimization of the project based on more detailed sub-factors.

For universities, this effective evaluation model can support and guide researchers in their enthusiasm and investment in technology transfer. Siegel et al. [50] proposed that universities play a crucial role in the distribution of benefits from research projects and the incentive system for researchers. Universities can ascertain their significance in technology transfer through such assessments. This support is essential throughout the technology transfer process. Moreover, due to its distinct nature, technology transfer is not a field of scientific research, but rather research applied to practical activities. Therefore, it requires a combination of knowledge from different fields. Universities should also enhance interdisciplinary collaboration and researchers’ understanding of business knowledge. Additionally, an effective technology transfer management system is conducive to improving the efficiency of Cooperative Technology Transfer. By evaluating the entire model, universities can identify the tasks of researchers at each stage and provide more targeted assistance.

For enterprises, their overall strength and willingness to cooperate are crucial. Companies should learn to select projects that align with their strengths. Many technology transfer projects collaborate with small companies because large enterprises have lost their original innovation capabilities due to their complex structure, bureaucratic procedures, and standardized work methods [73]. Enterprises should not rely solely on academic institutions but should actively engage in technology transfer partnerships. As entities with a deep understanding of the industry and market, enterprises should leverage their advantages to provide valuable insights into technology transfer activities. This is vital for the generation and absorption of technology. Furthermore, in most collaborative research models, the primary responsibility of the academic institution is to develop scientific and technological products, while the enterprise’s task is the management of technology transfer activities, and subsequently the employment of the relevant skills and experience to oversee the quality, cost, and schedule of the technology transfer.

6. Conclusions

This study supplements the research on CTT projects by examining the impact of cooperative relationship establishment, management factors, and technology achievement transformation. A comprehensive evaluation model with 13 influencing factors was established from the perspective of the three stages of CTT. The weights of the factors were determined by combining expert scores with an improved Gray Relation Analysis method. The obtained weights were then applied to the enhanced Dempster–Shafer evidence theory to evaluate the cases. Two cases in China were used to demonstrate the effectiveness of the model.

The case analysis demonstrates that CTT is based on the firm cooperation of enterprises and the strong research capabilities of R&D personnel. The attitudes of both the enterprise and the university not only determine the smooth development of the early-stage project initiation, but also provide necessary safeguards for potential project management issues that may arise later. However, medium-term management issues cannot be ignored, as poorly managed situations can lead to outcomes far beyond expectations. Despite the strong tolerance of both the enterprise and the university, these issues will inevitably bring potential risks.

The assessment model proposed in this paper serves as a valuable tool for exploring similar research questions and topics within the framework of cooperation between industries and universities. It emphasizes the importance of incorporating the expertise and insights of professionals in related investigations. Additionally, this project aids in understanding the challenges and lessons learned from the CTT framework, particularly as applied in an industrial country that has a significant need for more CTT projects.

However, this study lacks consideration of non-economic factors in CTT projects. This study found that the willingness and attitudes of both parties involved in CTT have a significant impact on the progress of the project, but the assessment mainly focuses on economic factors, lacking consideration of non-economic factors such as professor reputation and patent applications. Therefore, future research can build theoretical models in this direction to explore how to enhance the motivation for cooperation between universities and enterprises.

In addition, the proposed assessment model specifically targets CTT, while the other two models have their own unique functionalities. For instance, the direct transfer model needs to consider the technological capabilities of the company and the degree of alignment between technology and the market. On the other hand, the self-development model needs to take into account the economic capacity of the university. Therefore, when constructing the models, they differ based on the varying influencing factors. Also, these three models share some commonalities, such as evaluating the value of technological products. And it is necessary to further fine-tune this model in the future to incorporate assessment models for the other two technology transfer models.