Keywords

1 Introduction

The function of transmission is to transfer power. A transmission system refers to a complete set of devices or mechanisms that transfer power from a power source to an actuator. In the era of animal power and steam engines, there was almost no difference in speed between the power source output and the actuator output, so there was no need for a complex transmission system with deceleration. With the development of power sources from steam engines to IC engines, the output speed gap of IC engine and actuator increased, making it necessary to add deceleration between the power source and the actuator. As a result, transmission system with deceleration and other developed functions are widely used.

At early stage, people focused on the practical significance of transmission systems and the impact of the deceleration on the working performance of actuator. There were limitations on the layout space, structural shape, and weight of the deceleration mechanism between the power source and the actuator, and the feasibility of different deceleration mechanisms was also very important. At the same time, cost was also concerned. Gear ratio, performance, feasibility and cost became important indicators for the evaluation of transmission systems. Later on, the requirements for the handling experience of transmission systems and the auxiliary functions that could be achieved by transmission systems increased. The requirements for smooth speed switching, impact resistance, precise control, load holding capability, and power density were also introduced as evaluation indicators. In recent years, the development of safety, environmental protection, and intelligence has become deeply rooted. Young users in particular are looking for personalised experiences. Efficiency, energy saving, environmental friendliness, and the completeness of support facilities are becoming increasingly important. Environmental requirements and the uneven distribution of energy resources have elevated transmission efficiency to a key evaluation criterion in the process of selecting transmission technologies.

As technology evolves, there may be more complex requirements and functions that will be included in the evaluation indicators when considering different application scenarios, different market segments, and different customer groups.

In the current product development process, target functionalities and requirements are passed directly to design engineers from the relevant departments. Engineers do not analyse and prioritise the list of features and requirements. As a result, engineers may select the optimal technical solution based on the target functionalities and requirements, but this does not guarantee that the product will meet market and customer expectations. This can result in a very well-designed product that fails to gain customer acceptance in the marketplace [1], Fig. 1 try to state the relationship.

Fig. 1
A chart for the connections among the actors determining the selection of transmission systems and components. It includes speed ratio feasibility performance cost, power density impact resistance, precise control, energy saving environment, mature facility intelligentization, and high efficiency.

Connection of determining actors and customer satisfaction

Full size image

This article attempts to introduce the KANO requirements analysis model to the evaluation of functionality and requirements before product development. It also presents the application of three evaluation methods. By evaluating and filtering functionalities and requirements, the aim is to help engineers determine the reference indicators in engineering design and increase the success rate of projects.

2 Literature Review

The transmission system is a mechanical device that enables the transmission of power and motion. It acts as a connection between the power source and the executing mechanism. In the actual design process, there are many transmission solutions available to meet different customer needs.

Chen Yijun and Wu Kejian [2] considered how to select a transmission solution among numerous options for a machine. They found that applying fuzzy comprehensive evaluation theory to study systems composed of a limited number of components can effectively avoid the complexity and computational burden of traditional evaluation methods. The Analytic Hierarchy Process (AHP) [3], proposed by American operations researcher T L Saaty, is a method based on human judgement that can effectively quantify qualitative problems. It is concise, flexible, and logically strong, making it a feasible solution for multi-level, multi-objective optimization problems. Xiong Rui and Cao Kunsheng [4] introduced a decision-making method using “indicator matrices” to construct “judgment matrices” for solving quantitative or mixed-type multi-objective problems. To avoid the subjectivity and bias in the AHP method, Ma Yundong and Hu Mingdong [3] developed an improved Analytic Hierarchy Process called I-AHP, using the optimal transfer matrix. Song Baowei, Zeng Wenhua, Mao Zhaoyong, and Liang Qingwei [5] introduced a combined evaluation method that integrates the Analytic Hierarchy Process with fuzzy mathematics and incorporates fuzzy Analytic Hierarchy Process based on entropy weight, forming a subjective–objective combined evaluation method. Yao Huibo [6] studied four transmission scheme selection and evaluation models: fuzzy comprehensive judgment method, support vector machine method, uncertain measure model, and entropy weight-based fuzzy AHP method. He believed that the fuzzy comprehensive evaluation method can provide a relatively comprehensive quantitative analysis of all major influencing factors in the evaluation of mechanical transmission schemes, enabling objective and effective selection of more reasonable solutions.

As machinery and equipment are tools used to accomplish specific tasks, the transmission system is a part of the machinery or equipment. Functional requirements are defined by the tasks to be performed by the machinery and equipment and may have special characteristics due to different individual customer needs. The degree of success of a product which is reflected in customer satisfaction depends on the product's ability to meet actual functional requirements, and can only be achieved by meeting customer needs. If engineers rely on the target functional requirements provided by the relevant departments to complete the development and design, without verifying whether the target functional requirements match reality. Customer satisfaction will unable to be guaranteed, refer to Fig. 1.

According to Kelsey Miller's definition [7], customer needs are the problems that individuals are trying to solve, which drive them to seek products or services to address these problems. Customer needs include functional needs, social needs, and emotional needs. Jeffrey Bussgang [8] defines the consistency between a product and the target customer needs as product-market fit (PMF).

Without a method to analyse and classify target function before product development, it is inevitable that there may be discrepancies or deviations between the function provided by the product and the actual customer needs.

3 Solution

In the process of new product development, enterprises emphasize division of labour and collaboration, ask engineers designing products according to the requirements of the demand department. However, for some new products, it is not until the marketing stage that the market response is discovered and no product is sold, causing the new product programme to lose its intended value. Engineers may feel frustrated when they have found a solution based on reasonable evaluation methods but receive a negative market response. This situation may also lead to conflict with the collaborating departments. The product demand department expects engineers to design products that are closer to the market and more grounded.

The Fuzzy Analytic Hierarchy Process [5] based on entropy weighting can effectively solve the problem of selecting technical solutions at the evaluation level. However, it cannot weigh whether the function achieved by the right technical solution is the function required by the client. This indicates that product designers need to evaluates the product functional requirements before selecting technical solutions. Ranking product attributes based on how they are perceived by customers and their impact on customer satisfaction [9,10,11,12] by those who understand the technical options and functional requirements is an essential approach to success.

The Kano model [13, 17] established by Noritaki Kano which associates product design and production with customer demands. As a tool for enhancing products and services based on customer situations, the Kano model correlates customer satisfaction with product attributes. Based on the principle that different customer reactions are obtained depending on the degree to which the product meets their needs, the influencing factor attributes of the product are classified into five categories of customer satisfaction: attractive attributes (referred to as A attributes), one-dimensional attributes (referred to as O attributes), must-be attributes (referred to as M attributes), indifferent attributes (referred to as I attributes), and reverse attributes (referred to as R attributes).

  • Attractive attributes (A): the satisfaction level of users will not decrease if dysfunctional, but will greatly increase if functional.

  • One-dimensional attributes (O): the satisfaction level of users will increase if functional, but will decrease if dysfunctional.

  • Must-be attributes (M): When dysfunctional, customer satisfaction will significantly decrease.

  • Indifferent attributes (I): These are functions or needs that customers do not care about, and they have no impact on the user experience.

  • Reversal attributes (R): These are functions or needs that customers do not have at all. If functional, customer satisfaction will actually decrease.

The principles to be followed in product development and design are as follows: When designing a product, it is important to avoid indifferent attributes and reversal attributes as much as possible. After addressing the must-be attributes and one-dimensional attributes, efforts should be made to achieve attractive attributes for the product. This approach can effectively help engineers evaluate product requirements, obtain a combination of product requirements with higher overall customer satisfaction, and improve market fit.

There are two ways to collect customer feedback: by questionnaire or by customer interview, according to Sauerwein Elmar et al. [14], individual interviews seem to be more favourable, but interviews need to include questions for questionnaire and record properly. All the questions must be written clearly [16]. After collecting, the feedback of each customer must be sorted out with the following table.

Table 1 is one customer’s feedback, according to Griffin/Hauser, 20 to 30 customer interviews in homogenous segments is suffice to determine approximately 90–95% of all possible product requirements [14, 20]

Table 1 Satisfaction rating Table—The feedback of customer LI
Full size table

For multi-customer sample data, there are three methods for analysing and filtering customer feedback.

Discrete Analysis: In discrete analysis, the survey results are categorized based on the total count of attributes for each function or requirement. If there is a category with the highest frequency, the function or requirement can be classified into that category. If the results are close for different categories, they can be prioritized as follows [14]: Must-be > One-dimensional > Attractive > Indifferent. The importance of the function or requirement can also be determined by considering the feedback from the survey.

Quantitative Analysis (Better-Worse Coefficient): In this method, each function or requirement is classified based on its attributes, and the Better-Worse coefficient is calculated to represent its impact. The satisfaction coefficient when provided is calculated as Better/SI = (A + O)/(A + O + M + I) [14]. The Better/SI value is usually positive, indicating that satisfying a particular function or requirement will increase satisfaction. The closer the value is to 1, the stronger the impact of that function or requirement on satisfaction, making the product more favourable. The dissatisfaction coefficient when not provided is calculated as Worse/DSI = -1*(O + M)/(A + O + M + I) [14]. The Worse/DSI value is usually negative, indicating that not meeting a certain function or requirement will decrease satisfaction. The closer the value is to -1, the greater the reverse impact of that function or requirement on satisfaction, suggesting it should be included in the standard configuration.

Continuous Analysis: In continuous analysis, the results are converted into scores ranging from -2 to 4 based on the level of functional [15, 16, 18, 19]:—2 (I dislike it that way), -1 (I can live with it that way), 0 (I am neutral), 2 (It must be that way), 4 (I like it that way) when functional. -2 (I like it that way), -1 (It must be that way), 0 (I am neutral), 2 (I can live with it that way), 4 (I dislike it that way) when dysfunctional. A higher score reflects a greater preference for that function by customers.

3.1 An Example

Collected 30 feedback results from customers, for the Needs XI, by attribute evaluation: 15 evaluation bit A, 6 evaluation bit O, 6 evaluation bit I, 3 evaluation bit M, the details in the following label.

Then, based on the data of Table 2, the results of the three analytical methods will be.

Table 2 Satisfaction Rating Table—The feedback of Customer 1 ~ Customer 30
Full size table

The result of Discrete Analysis is presented in Table 3.

Table 3 Results of discrete analysis
Full size table

The results of Quantitative Analysis.

Satisfaction coefficient while functional: better/SI = (A + O)/(A + O + M + I) = (15 + 6)/ (15 + 6 + 6 + 3) = 0.7, which is close to 1 and has a stronger impact.

Dissatisfaction coefficient while Dysfunctional: Worse/DSI = -1*(O + M)/(A + O + M + I) = -1*(6 + 6)/ (15 + 6 + 6 + 3) = -0.4, which is far away from -1 and has a weak influence.

The Better-worse coefficients are more consistent with the characteristics of the desired attribute (O).

The results of Continuous Analysis.

Functional: 15*4 + 6*4 + 6*2 + 3*0 = 96, which is 80% of the total score of 120(=4*30).

Dysfunctional: 6*(-1) + 3*0 + 3*0 + 6*2 + 3*2 + 6*4 + 3*4 = 48, accounting for 40% of the total score of 120 (=4*30).

If 50% is identified as the cut-off point for strong or weak influence, then influence is stronger when functional and weaker when dysfunctional. It is more consistent with the desired attribute (O) characteristic.

3.2 Discussion on the Results

Among the three analysis methods, the result of the Discrete Analysis is A, while the results of the Quantitative Analysis and the Continuous Analysis are both O. The results obtained from these three methods show discrepancies. According to the KANO model, category A represents “the satisfaction level of users will not decrease if dysfunctional, but will greatly increase if functional,” while category O represents “the satisfaction level of users will increase if functional, but will decrease if dysfunctional.” These two classifications have different impacts on customer satisfaction. Therefore, it is necessary to use the KANO model to analyze all the functions or needs of the product again.

Among all feedback A in Table 2, there are various scenarios ranging from “ It must be that way” to “I dislike it that way” for the feedback if dysfunctional. It is evident that the categorization of all these results as category A lacks sufficient discrimination.

The results of the Quantitative Analysis and the Continuous Analysis are almost similar, but the Quantitative Analysis also suffers from the same problem as the Discrete Analysis. The Continuous Analysis scores the different feedback results in category A from two dimensions, making the results more reliable.

4 Conclusion

Based on customer’s different personalized understanding of functions or needs, an example of function and need analysis using the KANO model shows that it is crucial to analyze and select the set of functions or needs for a product before its development. This analysis is expected to be beneficial in effectively utilizing resources to prioritize key functions or needs. By using the KANO model, the functions or needs of a product can be analyzed based on customer feedback. Among the three analysis methods, the Continuous Analysis has a higher ability to differentiate customer feedback results, making it more advantageous in obtaining reliable results.