A BP Neural Network Product Design Optimization Model Based on Emotional Design and Sustainable Product Design

Abstract

The objective of this study is to investigate the impact of emotional design and sustainable product design on user experience and satisfaction. Additionally, the study aims to optimize the combination efficiency and material selection methods of these two design approaches, extracting optimization parameters based on emotional and sustainable product design principles. By analyzing user satisfaction through questionnaires, various product indicators were examined. The study ultimately establishes a BP neural network optimization model by analyzing the design elements of different products and the satisfaction levels of various indicators. This BP neural network design optimization model, integrating emotional and sustainable design, demonstrates an accuracy rate of over 95% in predicting user satisfaction for different design schemes. Upon examining the emotional design and sustainable design parameter tables following data standardization, and analyzing various product models, it was concluded that the rate of change in each design parameter significantly impacts the satisfaction of each product indicator, highlighting the importance of these parameters in product design optimization.

1. Introduction

1.1. Background and Purpose

In recent years, there has been a gradual shift in focus towards sustainable development, environmental protection, and the promotion of a green economy [1,2]. Furthermore, the problems of resource shortages and environmental pollution caused by industrial development have also gradually emerged [3]. The advent of ISO 26000 [4] has prompted the international community to actively promote sustainable design management [5]. Currently, all industries are engaged in the exploration and pursuit of sustainable development models. Among these, product design is closely connected to people’s everyday lives, and integrating the concept of sustainable development into product design can better align with the current core trend of development in the field of industrial design [6,7,8]. The replacement of materials during the iteration of a product will result in a different user experience. In the context of sustainable product design, the utilization of sustainable materials is frequently regarded as a pivotal factor [9,10]. There is a discernible correlation between the minimalist design style in product design and sustainable development, characterized by a commonality of moderation, essence, and efficiency [11]. Nevertheless, it appears that utilizing sustainable materials alone in product design is unable to fulfill the user’s innate desire for product appearance, their behavioral need for user experience, and their spiritual need for environmental protection [12]. The challenge for product designers today is to select suitable sustainable materials for product design and production while also satisfying the user experience and achieving a balance between the usability and aesthetics of the product [13]. In the context of new sustainable materials, designers tend to rely on past design experience and industry standards for design attempts [14], which is not only inefficient but also likely to reduce the user experience due to the material performance of the designed product. Consequently, it is crucial to employ scientific methodologies to optimally integrate emotional design and sustainable product design.

The objective of this paper is to propose a design model based on the BP neural network, which combines emotional design and sustainable product design. This model aims to provide a theoretical and factual basis for the selection and use of sustainable materials in future sustainable product design.

1.2. Methodology and Significance

This study employs both survey research and experimental research methodologies [15]. Firstly, we utilize Apple’s products, which actively utilize sustainable materials in the field of sustainable development and have a wider audience as the starting point. We then exclude the influence of the software level on the user experience and analyze and extract the emotional factors of affective design and the elements of sustainable product design for the use of materials and the appearance design level of the products [16]. Finally, we collect the relevant performance data of the representative products and then conduct user research for each representative product by means of questionnaire surveys. Subsequently, the performance data of the representative products are collated, and user research is conducted on each representative product via questionnaires. The BP neural network model is constructed based on the actual data of the products and the user research data. Once the model is constructed, the corresponding data are selected for verification purposes.

The objective of this study is to develop an algorithmic model that integrates emotional design and sustainable design principles. In the future, when designers are faced with new sustainable materials, they will be able to replace the relevant product parameters and user research data in a specific field to derive the product design parameters and approximate user satisfaction in line with the field. This will enable the application of the derived parameters to a wider range of scenarios with higher efficiency.

2. Literature Review

2.1. Emotional Design

The concept of emotional design was first proposed by Donald Norman in the late 1980s, according to the three-level theoretical model of emotional design (as shown in Figure 1). This model divides the emotional needs of users into three layers: visceral, behavioral, and reflective [17]. Emotional design is a product design approach that prioritizes the emotional needs of users in order to optimize the user experience [18]. Emotional design is a methodology for translating human emotions into product design, integrating both aesthetic considerations and the user experience [19]. In accordance with this concept, the objective of design encompasses the emotionalization of product form, product characteristics, and operation. This encompasses the integration of design psychology, aesthetics, economics, and other multidisciplinary fields [20]. It can be integrated with multiple disciplines.

The product is an object that evokes emotions in the user [21,22], and emotional product design is a design method that begins with the user’s internal and external needs [23]. The implementation of emotional design in product design is primarily concerned with the form, material, color, and other aspects of the product, which collectively influence the user’s experience [24]. Consequently, the equilibrium between product aesthetics and experience can be attained.

2.2. Theories Related to Sustainable Design

2.2.1. Sustainable Design

The concept of sustainable design is still a matter of debate within the academic community [25]. However, there is a growing consensus that sustainable design is closely related to the concepts of green design, eco-design, design for the environment, and life-cycle design [26]. Sustainable design is a strategic design activity that aims to reduce environmental pollution and resource consumption by integrating products and product services to build and develop sustainable solutions [27]. In the context of sustainable design, it is essential to consider the economic, environmental, and social aspects of the design process and an integrated approach is required to balance the needs of these areas [28]. This represents a kind of recycling design thinking, and the concept also encompasses social and cultural sustainability [29], which is shown in the conceptual relationship diagram in Figure 2. As a forward-thinking, innovative concept that transcends the narrow boundaries of traditional design, designers consider the practicality and aesthetics of products while working to minimize energy consumption and the environmental impact of the production process, thereby promoting sustainable development [30]. The concept of sustainable design is of great significance to humanity’s future development.

2.2.2. Product Design in the Context of Sustainability Theory

In the past, product design was primarily concerned with meeting the current needs of humans, with success being measured in economic terms [31]. In response to the environmental problems caused by rapid product iterations, sustainability theory has been integrated into product design [32]. In the context of manufacturing companies, the core concept of green design is typically encapsulated by the “3R” principle, which encompasses “Reduce, Reuse, and Recycle” [33]. The concept of sustainable design can be reflected in product design through the structure of the product, the choice of materials, and the loading and unloading of the product. Among these, the selection and application of materials that are sustainable can most easily and intuitively distinguish whether a product is sustainable or not. Furthermore, the selection of recycled materials for reprocessing can also save raw materials and reduce production costs.

2.3. BP Neural Network

The BP neural network (back propagation neural network) is based on the multilayer perceptron model. The introduction of nonlinear function activation, together with the direction of the algorithm to optimize the neural network [34], constitutes the artificial neural network. The BP neural network is one of the most widely used neural network models in the field of artificial neural networks (ANNs) [35]. It is commonly used in the field of data prediction, as it is a multilayer feed-forward neural network that can be applied to both linear and nonlinear datasets [36]. Furthermore, it can accurately predict unknown scenario data by learning the mapping relationship between input and output data [37]. In the field of product design, the BP neural network is currently employed in the prediction and evaluation of product innovation programs based on users’ perceptual intention demand [38]. This is carried out to reduce market risk and improve product competitiveness.

3. Constructing Optimization Algorithms Based on Emotional Design and Sustainable Product Design Ideas

A new material design analysis and prediction model was established based on a BP neural network model, which is guided by emotional design and sustainable product design. This model is called the BP neural network design parameter optimization model, and it is based on emotional design and sustainable design. The extraction of emotional design features and sustainable material features as analysis elements; the analysis of user satisfaction with various indicators of the product; and the establishment of a dataset to train the BP neural network design parameter optimization model based on emotional design and sustainable design are all key steps in this process. The objective is to investigate the impact of emotional design and sustainable design indicators on user experience. This is achieved by calculating the user’s satisfaction with various product aspects using a model that relies on the extraction of material characteristics and design elements. The model then provides feedback to designers, enabling them to further optimize the product design based on the user’s satisfaction.

3.1. Example Analysis

The rationale for selecting Apple products as the subject of research into emotional design and sustainable product design algorithm modeling is that Apple is a relatively mature player in the field of sustainable development on a global scale and has achieved notable results. The sustainable design concept, which incorporates emotional design, has had a profound impact on the entire industry and users around the world. Although Apple’s design is known for its simplicity and minimalism, which can reduce users’ emotional arousal regarding the appearance, the nearly identical minimalist design and the adoption of the same type of operating system can eliminate the subjective influence of users’ preferences on different appearance design schemes and different operating systems on product satisfaction in the experimental data. Concurrently, Apple’s multi-generation products are designed and produced with recycled aluminum metal. However, there are aluminum alloys with varying ratios, and there are also products of the same size and different materials in the side-by-side comparisons. Therefore, it is convenient to analyze the emotional and sustainable design elements of the product’s appearance by using Apple’s products as the subject of experimental data collection.

3.2. Emotional Design Feature Extraction

The field of emotional design is of great importance in the context of product design. Its objective is to establish an emotional connection between the user and the product, with the aim of providing and enhancing the user experience. The extraction of emotional design elements should be based on the three levels of emotional design requirements, namely visceral, behavioral, and reflective. The design of the visceral layer is contingent upon the product’s visual appeal, tactile properties, and material composition. The manner in which a product is used can influence the preferences and acceptance of its appearance. For instance, in a domestic setting, users may be more inclined to prioritize the color, design, material, and appearance of a product. Conversely, in an outdoor environment, users may favor product designs with durable, waterproof, and portable attributes. The behavioral layer of design is concerned with the ease of use and functionality of the product. The usage scenario affects the user’s expectations regarding the functionality and operation of the product. For instance, in the context of mobile scenarios, users prioritize a comfortable grip in product design, whereas, in stationary office scenarios, users tend to place greater emphasis on the wider field of view for multitasking afforded by larger screen sizes. By developing a comprehensive understanding of the usage scenarios, designers can enhance the functionality and operational flow of the product and improve the ease of use and practicality of the product. The reflective layer is concerned with the impact of the product on the user’s life. This prompts the user to consider the product experience in a specific scenario, thereby facilitating a form of self-identification [39]. In accordance with the three levels of emotional design, the emotional design parameters of a product can be extracted. The elements of the visceral layer are the color of the product and the appearance of the sustainable materials used. The behavioral layer encompasses elements such as the screen size and weight of the product. The reflective layer elements are usage scenarios.

The term “usage scenario” is used to describe the specific environments and situations in which the user interacts with a product. In contrast, the concept of “emotional design” refers to a design method that aims to evoke a specific emotional response from the user through the strategic use of design elements, ultimately enhancing the overall user experience. A usage scenario can elucidate the functional requirements. Different usage scenarios give rise to disparate functional requirements. Furthermore, different usage scenarios can influence the user’s preference and acceptance of the product’s appearance, as well as their expectations regarding the product’s functions and operational methods. Consequently, it is possible to circumvent the investment of resources in superfluous functions and designs and instead concentrate on the pivotal features pertinent to specific scenarios. This approach enhances the efficiency of the design process and reduces production costs. The definition of product usage scenarios can enhance product adaptability. Different scenarios may exist in different environmental conditions. For instance, portable product design must consider the daily use of inadvertent falls and the mobile carrying process caused by external forces and bending phenomena. It is possible to determine the usage scenario in order to enable the design program to take these factors into account, thereby improving the adaptability and stability of the product in a variety of environments. Furthermore, by accurately matching the design to the user’s scenario, it is possible to provide a more intimate and satisfactory experience. The determination of the intended usage scenario represents a pivotal stage in the process of electronic product design. This stage can have a significant impact on the overall quality, practicality, and market value of the product.

The impact of product size design on the user can be analyzed from three levels: ergonomics, the emotional resonance of the user, and cultural considerations. From an ergonomic perspective, the product size can be designed to align with the hand size and usage habits of the majority of users, enhancing comfort and convenience. From an emotional resonance standpoint, size design can be employed to convey the product’s personality and values. For instance, a compact product may be perceived as conveying qualities of sophistication and portability, whereas a large product may be associated with its function and strength. In terms of cultural considerations, users from different cultures and geographical regions have disparate preferences and expectations regarding size. Consequently, these variables must be considered to develop an effective size strategy for the global market.

The design of screen size can influence users’ choice of products on three levels: the visual experience, content presentation, and emotional connection. From the perspective of visual experience, screen size has a direct impact on the user’s visual experience. An appropriate screen size can ensure that users can comfortably view and operate any content, thereby avoiding eye fatigue. In terms of content presentation, the appropriate screen size is selected to optimize the presentation of the content, taking into account the main functions of the product and the type of content. For instance, in products where reading or watching videos is the primary function, a larger screen may be required to provide a superior experience. In terms of emotional connection, the design of the screen is employed to enhance the emotional connection between the user and the product. For instance, a screen with high resolution and vibrant colors can facilitate a more pleasurable visual experience.

The product color scheme represents the primary manifestation of emotional design, which influences customer decision-making in three distinct ways: through the elicitation of psychological feelings, the provocation of emotional responses, and the shaping of behavioral decision-making. The application of color schemes can elicit a range of psychological responses in users. For instance, warm colors can elicit positive emotions and a sense of vitality, whereas cool colors (such as blue and green) can evoke a sense of tranquility and calmness. Designers must select the most appropriate color scheme for a given product and target user groups in order to create an atmosphere that aligns with the psychological expectations of users. Moreover, the color scheme can influence the emotional response of the user. The application of different color schemes can elicit a range of emotional experiences, including excitement, pleasure, and relaxation, among others. In product design, designers will make use of the principle of color psychology through the color scheme to influence the user’s emotional response, with the intention of enhancing the emotional connection between the user and the product. In addition, the color scheme will also affect the user’s behavioral decisions. For instance, in website design, different color schemes can direct the user’s gaze and attention and influence the user’s browsing behavior and click rate. Designers will utilize the color scheme to highlight the crucial information and functional aspects of the product, thereby influencing the user’s purchasing and usage decisions.

The field of emotional design is of significant importance in the realm of product design. The manner in which the size and color schemes of products are employed can have a profound impact on users, particularly when these elements are based on the visceral layer. Appropriate product size can enhance the user’s sense of grip and ease the operation and portability of the product, as well as its aesthetics and practicality. Conversely, the appropriate color scheme can trigger different psychological feelings, emotional responses, and behavioral decisions. Consequently, in the process of product design, it is imperative that designers consider the emotional needs and psychological expectations of users and enhance the emotional connection between users and products through reasonable product size and color schemes. This will improve users’ experience and satisfaction.

3.3. Sustainable Product Design Materials Feature Extraction

The application of sustainable materials can directly influence the sustainability of products. This study primarily examines the impact of sustainable design on product design, with a particular focus on the role of sustainable materials. The hardness, density, and strength of sustainable materials are crucial parameters in product design, exerting a profound influence on the performance, lifespan, cost, and user experience of products.

The impact of material hardness on products can be observed in four key areas: product touch and comfort, wear resistance and durability, structural stability, and processing difficulty. With regard to product touch and comfort, hardness has a direct impact on the tactile experience of the product. For instance, the hardness of a key can influence the user’s operating comfort and perception of the product. In terms of wear resistance and durability, products with higher hardness have a more wear-resistant surface and are suitable for applications that require long-term friction or impact, such as mechanical parts and tools. The impact of structural stability on the product is reflected in the high hardness of the material, which tends to have better structural stability. This helps to maintain the shape and dimensional accuracy of the product. Concurrently, the hardness of the material and the difficulty of processing will influence the cost of the product. The necessity of higher cutting and grinding forces during the processing of harder materials may result in increased difficulty and higher costs.

The impact of material density on the product is primarily manifested in the weight, volume, and quality of the product, as well as the associated cost and resource consumption. The most immediately apparent impact of material density on the product is reflected in the weight. Density directly affects the weight of the product, which, in turn, affects the handling, installation, and use of the product. In instances where a lightweight design is required, the selection of a less dense material can assist in reducing the weight of the product. Furthermore, material density affects the volume–mass relationship of a product. For a given volume, the greater the density of the material, the greater the mass, which affects the stability and load-carrying capacity of the product. Finally, and most central to the concept of sustainable design, is the issue of cost and resource consumption. The production of a given volume of a product from a denser material will require a greater quantity of raw materials, which can result in higher costs and increased resource consumption.

The impact of material strength on products can be observed in three distinct areas: load-bearing capacity, safety and reliability, and structural design. Among these factors, the load-bearing capacity is reflected in the high strength of the material. Products with greater load-bearing capacity can withstand greater loads and impacts. This is applicable to the need to withstand heavy loads or high-impact occurrences, such as building structures, automotive parts, and so on. In terms of safety and reliability, products with high material strength are less susceptible to damage or failure when subjected to external forces, which enhances the safety and reliability of the product. In the product design process, the appropriate strength index is selected according to the intended use of the product and the load requirements. This is then reflected in the corresponding structural design, ensuring that the strength of the product meets the required specifications.

In conclusion, the parameters of hardness, ductility, density, and strength occupy a pivotal position in the product design process. In the context of product design, it is essential to consider the impact of these parameters in a comprehensive manner, with a view to selecting and optimizing them in a manner that is aligned with the specific needs of the product in question. Figure 3 shows the flowchart of the BP neural network design parameter optimization model based on emotional and sustainable design.

3.4. Optimization Algorithm Creation

The design of the dimensions and color schemes in terms of emotional design can be visualized in the product itself. In the context of sustainable design, the properties of sustainable materials can also be incorporated into the product characteristics to a certain extent. For instance, the density of the material will be reflected in the weight of the product in accordance with the alteration of the product size in the context of emotional design. Additionally, the strength of the material will also influence the bending resistance of the product, which may consequently impact the user experience to a certain extent.

The objective of this paper is to construct an optimization algorithm that explores the correlation between sustainable design and emotional design. When a designer encounters a new material design, they can analyze the user’s use of the scenario and the material characteristics to derive a preliminary design plan. The model can then analyze the designer’s design plan to predict the user’s satisfaction with product indicators. Additionally, the model can provide the value of the design characteristics of the product, offering guidance to the designer. The model is capable of analyzing the designer’s design scheme and predicting the user’s satisfaction with various product indicators. Additionally, it can identify the influence of various design features on the product, thereby providing the designer with guidance regarding potential avenues for further optimization of the product design.

In order to investigate the influence of emotional design and sustainable design on user experience, this study identifies the following analytical elements in emotional design: product length, product width, product thickness, screen size, product weight, and color scheme. Additionally, the following analytical elements are extracted for sustainable design: material density, tensile strength, yield strength, material hardness, and material stiffness. These elements are then combined to create a product dataset (as shown in

Table 1. Product datasets.

Model	Size	Weight	Feel Drop	Resistance	Bending Resistance	Color
3gs	2	2.2	3.4	2.5	2	1.2
4	2	2.7	3.2	1.7	1.4	2.3
5	2.2	2.1	2.9	2.5	1.4	1.2
5c	2.3	3.1	3.6	2.8	2.1	3.9
5s	2.9	2.5	3.1	2.6	2.4	3.6
6	3	2.8	4.1	2.1	1.8	3
6p	4.2	3.2	3.4	2.9	2.1	3.1
8	2.9	2.7	3.3	3.4	3.5	3
8p	3.4	2.9	3.3	2.9	3.2	3.7
x	3.4	2.9	3.9	3.2	3.3	2.3
xs max	3.5	3.6	3.1	3.6	3.6	2.8
xr	3.1	3.6	3.1	3.2	3.2	3.9
11	3.5	2.9	3.6	3.4	3.1	4.2
11p	3.2	3.5	3.1	3.2	3.2	2.4
11pm	4.2	3.7	3	3.6	3.3	2.4
12	3	2.7	3.2	3	3.3	3.8
12mini	2	2.9	3.8	3.2	3.4	3.7
12pro	3.5	3.5	2.5	3	3	2.6
12pm	4.5	3.9	3.3	3.6	2.8	3
ipad1	3.2	3.2	3.3	3.2	3.3	1.2
ipad p9.7	2.5	3.1	3.5	2.9	2.8	2.4
ipad mini	2	2.7	3.3	2.7	2.8	1.8
ipad p11	3.1	2.9	2.6	2.7	2.1	2
ipadp12.9	4.3	4.4	2.2	2.5	2.3	1.6
mba11	2.8	2.7	2.9	2.5	2.4	1.7
mba13	2.7	1.6	3.9	3.3	3.1	4
mbp14	3.2	3.4	3.5	3.6	3.7	2.8
mbp15	3.9	3.3	3.4	3.3	3	2.2
mbp16	4.1	3.9	3.8	3.4	3.6	2.6
mbp13	2.7	3.3	3.6	3.2	3.1	1.6

).

3.5. Predictive Modeling Creation

The above three dimensions of emotional design select product length, product width, product thickness, screen size, product weight, color scheme, and product use environment as analysis elements and material density, tensile strength, yield strength, material hardness, and material stiffness as analysis elements in terms of sustainable design. Next, we need to analyze the impact of these design elements on the product and the feedback from users on these parameters.

There is a discernible correlation between the impact of the design parameters selected for model construction and the product. The dimensions of the product, including the length, width, height, and material density, as well as the optional screen size, will influence the product’s weight. The material’s tensile strength, yield strength, hardness, and stiffness will determine the constraints of the product’s dimensional design, as well as its bending and drop strength. In the context of sustainable design, a stronger product can ensure its performance. In the context of sustainable design, enhanced product strength ensures a longer product lifespan. The dimensions, mass, and material properties of the product can also influence the user’s grip. In this context, the selected output values of the model are users’ satisfaction with the product’s screen size, weight, drop resistance, bending resistance, and color. Consequently, designers can obtain user feedback on the design scheme through the output value of the model.

4. Analysis and Discussion

4.1. Experimental Methods

The BP neural network, or error feedback neural network algorithm, is a multilayer feed-forward neural network that has been trained according to the error back propagation algorithm (as shown in Figure 4). The structure of the network is straightforward and comprises three principal layers: the input layer, the hidden layer, and the output layer. The input layer is the initial layer of the neural network that receives external information. It contains multiple nodes, each representing an input feature or variable. The hidden layer constitutes the central component of the neural network, situated between the input and output layers. The nodes in the hidden layer are not directly connected to the external environment, yet their state can influence the relationship between the input and output. The output layer is the final layer of the neural network and is responsible for generating the output of the neural network, which may deviate from the desired result of the actual problem. The BP neural network adjusts the network parameters to minimize this error through the error back propagation algorithm. This experiment is based on the dataset established in Section 3.4 (Table 1), which extracts the product’s affective design parameters and sustainable design parameters as inputs for training the model. The various satisfaction levels of the product serve as the model’s outputs. The corresponding user satisfaction is used as output values. In order to eliminate the effect of the difference in dimension between the input and output values on the accuracy of the model, data normalization is employed in scikit-learn. This involves the normalization of all the input and output data. The design parameters derived from affective design and sustainable design are inputted into two fully connected layers. The number of neurons in the first layer is 256, while the second layer contains 512 neurons. The number of neurons in the output layer is equivalent to the number of labels, specifically 10 neurons for the design parameter and 10 neurons for the user satisfaction parameter. The data in Table 1, which combine the design parameters extracted from different products based on emotional design and sustainable design with user satisfaction, were selected for the model dataset. Subsequently, 80% of the data were randomly selected as the training set of the model, which was put into the BP neural network prediction model for training. The remaining 20% of the data were not involved in the training and served as the validation set, which was used for validating the results of the model.

The BP neural network comprises a hidden layer, which contains a neuron structure. The neurons in each layer are connected to each other by weights. The weights are adjusted continuously during the neural network training process in order to optimize the network’s performance. The activation function is employed to introduce nonlinearities in the neurons, thereby enabling the neural network to address complex nonlinear problems. The training process of the BP neural network comprises two phases: forward propagation and back propagation. In the forward propagation phase, the input signal acts on the output node via the hidden layer, undergoing a nonlinear transformation to produce an output signal. In the event that the actual output does not correspond to the desired output, the back propagation stage is initiated. This involves the back propagation of the output error layer by layer through the implicit layer to the input layer. The network parameters (e.g., weights and biases) are then updated in order to reduce the error. X_n represents the nth design parameter, and W_n (weight) is the corresponding weight of X_n, which represents the magnitude of the effect of X_n on the next neuron. For a single neuron, the positive feedback signal it obtains is the sum of the previous values of each node and its weight.

$$\mathrm{n}\mathrm{e}\mathrm{t}=\sum _{\mathrm{i}=0}^{\mathrm{n}}({\mathrm{w}}_{\mathrm{n}}{\mathrm{x}}_{\mathrm{n}}+{\mathrm{b}}_{\mathrm{n}})$$

The objective of this analysis is to determine the impact of the data from each time node on the prediction outcomes. Once a neuron has received a positive feedback signal, it must then apply the activation function to this signal before it can be output as the positive feedback signal of the next neuron.

$$\mathrm{n}\mathrm{e}\mathrm{t}\_\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{p}\mathrm{u}\mathrm{t}=\mathrm{R}\mathrm{e}\mathrm{l}\mathrm{u}\left(\mathrm{n}\mathrm{e}\mathrm{t}\right)$$

In this experiment, the Relu function will set the values below zero to zero, thereby reducing the influence of user satisfaction with large personal factors in the model. Once these steps have been completed, the discrepancy between the experimental design and the associated numerical error can be effectively addressed.

Prior to the data entering the BP neural network model training, data normalization is performed (as shown in

Table 2. Design parameters after data standardization.

	Emotional Design						Sustainable Product Design
Model	Fuselage Width	Fuselage Thickness	Screen Size	Product Weight	Carrying or Not	Product Color Categories	Material Density	Tensile Strength	Yield Strength	Material Hardness
3gs	0.18	1.00	0.00	0.20	0.00	0.00	0.00	0.00	0.00	0.02
4	0.00	0.44	0.00	0.22	0.00	0.00	1.00	0.26	0.14	1.00
5	0.00	0.13	0.16	0.00	0.00	0.00	0.23	0.26	0.14	0.00
5c	0.03	0.38	0.16	0.18	0.00	0.75	0.00	0.00	0.00	0.02
5s	0.00	0.13	0.16	0.00	0.00	0.25	0.23	0.26	0.14	0.00
6	0.43	0.00	0.38	0.15	0.00	0.25	0.23	0.26	0.14	0.00
6p	0.98	0.04	0.63	0.53	0.00	0.25	0.23	0.26	0.14	0.00
8	0.45	0.07	0.38	0.32	0.00	0.50	0.24	1.00	1.00	0.69
8p	1.00	0.11	0.63	0.79	0.00	0.50	0.24	1.00	1.00	0.69
x	0.63	0.15	0.72	0.54	0.00	0.00	1.00	0.26	0.14	1.00
xs max	0.96	0.15	0.94	0.84	0.00	0.25	1.00	0.26	0.14	1.00
xr	0.88	0.26	0.81	0.72	0.00	1.00	0.24	1.00	1.00	0.69
11	0.88	0.26	0.81	0.72	0.00	1.00	0.24	1.00	1.00	0.69
11p	0.66	0.22	0.72	0.67	0.00	0.50	1.00	0.26	0.14	1.00
11pm	0.98	0.22	0.94	1.00	0.00	0.50	1.00	0.26	0.14	1.00
12	0.66	0.09	0.81	0.44	0.00	1.00	0.24	1.00	1.00	0.69
12mini	0.29	0.09	0.59	0.18	0.00	1.00	0.24	1.00	1.00	0.69
12pro	0.66	0.09	0.81	0.66	0.00	0.50	1.00	0.26	0.14	1.00
12pm	1.00	0.09	1.00	1.00	0.00	0.50	1.00	0.26	0.14	1.00
ipad1	0.64	1.00	0.36	0.91	1.00	0.00	0.00	0.00	0.00	0.00
ipad p9.7	0.41	0.03	0.36	0.28	1.00	1.00	0.00	0.00	0.00	0.00
ipad mini	0.00	0.21	0.00	0.00	0.00	0.67	0.00	0.00	0.00	0.00
ipad p11	0.51	0.00	0.62	0.36	0.00	0.33	1.00	1.00	1.00	1.00
ipadp12.9	1.00	0.13	1.00	1.00	0.00	0.67	0.00	0.00	0.00	0.00
mba11	0.00	0.11	0.00	0.00	1.00	0.00	0.00	0.00	0.00	0.00
mba13	0.36	0.00	0.18	0.13	1.00	1.00	0.00	0.00	0.00	0.00
mbp14	0.52	0.55	0.24	0.44	1.00	0.33	1.00	1.00	1.00	1.00
mbp15	0.98	1.00	0.35	0.81	0.00	0.00	0.00	0.00	0.00	0.00
mbp16	1.00	0.79	0.42	1.00	0.00	0.33	1.00	1.00	1.00	1.00
mbp13	0.48	1.00	0.15	0.42	1.00	0.00	0.00	0.00	0.00	0.00

), a technique commonly used in the data preprocessing stage to transform the data to a specific range, usually [0, 1] or [−1, 1], or to ensure that the data have unit variance and zero mean. This treatment is of significant importance for numerous machine learning algorithms and statistical analysis methods, as it eliminates scale discrepancies between features, ensuring that they contribute equally to the mode.

4.2. Model Results Presentation

Figure 5 illustrates the outcomes of the training of six sets of product data with distinct models utilizing BP neural networks. The mean square error (MSE) of the final model for the training set is recorded as Train ERROR = 2.8%, and that for the test set is recorded as Test ERROR = 4.3%. These values indicate that the prediction model has an accuracy of more than 95%. The training models depicted in Figure 5 employed data normalization to eliminate the differences in dimension between the individual design parameters, as well as the introduction of the activation function Relu to enhance the nonlinear fitting ability of the model. Observing the images reveals that the predicted values for product satisfaction derived from the MLP model exhibit a similar trend to the actual values. This indicates that the model has successfully derived the relationship between the influence of each design parameter on product form. Furthermore, it demonstrates that the problem of the change in user satisfaction between the various indicators of the product caused by the change in the different design parameters has been solved. This provides evidence of the model’s feasibility to optimize the design parameters based on the combination of a BP neural network and sustainable design. The feasibility of a design parameter optimization model based on a BP neural network combined with sustainable design was evaluated.

4.3. Model Results Presentation

The results of the user satisfaction prediction model that combines emotional design and sustainable design indicate that the prediction model can accurately predict user satisfaction using various product indicators under the design scheme. Furthermore, the mean square error (MSE) of the test set of the prediction model was recorded as Test ERROR = 4.3%, which indicates that the accuracy of the prediction model is more than 95%. This suggests that the prediction model can effectively guide designers in optimizing product design.

4.3.1. Analysis of the Results of the Sustainable Design Parameters

In the process of building the prediction model, the sustainable designers chose density, hardness, and strength as the design parameters of sustainable design in terms of materials. The density of the material selected during the product design process affects the weight of the product, which in turn affects the product’s grip; the hardness determines the product’s drop resistance, and the strength of the material determines the product’s torsion resistance, which ensures the product’s service life. At the same time, in order to ensure that there is no excessive involvement of the user’s emotional level, this side of the selection of cell phone product lines with mobile attributes is the main element of analysis (as shown in Figure 6)

For this analysis, the 3GS, 4, 5, 6, 6 Plus, 8, and 8 Plus, as well as the full range of 11th and 12th generation products, were selected as the targets for analysis. The first step is to determine the product usage environment for this controlled experiment as a mobile scenario. In this usage scenario, according to the hot zone diagram in Figure 6, the user data in terms of sustainable design parameters are positively correlated with product satisfaction. Regarding the analysis of material density and user satisfaction with product weight: material density is positively correlated with the weight of the product. The model’s output indicates that user satisfaction with product weight does not always favor lighter products, which shows that the impact of design parameters on the product is not a simple one-to-one mapping. This further illustrates the effectiveness of the BP neural network prediction model. For sustainable design, balancing product strength with thin and light design is a significant challenge. In product design, one cannot simply pursue thinness and lightness at the expense of the product’s strength. In the experimental situation with various materials to choose from, take the 6 Plus and 8 Plus for example. Despite having nearly the same size design, the 8 Plus uses a stronger back plate material, addressing the bending issue present in the 6 Plus. As a result, the overall user satisfaction is notably higher for the 8 Plus. In the limited material experimental environment, a horizontal comparison of the same material in the 6 and 6 Plus models reveals that the pursuit of a thin and light design in the 6, at the expense of product strength, results in lower user satisfaction with the product’s strength compared to the 6 Plus. These issues are reflected across the 11th and 12th generation products.

4.3.2. Analysis of the Results of the Emotional Design Parameters

In the process of developing the predictive model, the intuitive, layer-based elements of the sustainable design aspect were the color and appearance of the product and the materials used. The behavioral layer encompasses two elements: screen size and product weight. The reflective layer elements were usage scenarios. In the analysis of emotional design elements, the Mac series product line was employed, with the distinction of usage scenarios (as shown in Figure 7).

Firstly, from the perspective of usage scenarios, this model analyzes the experimental environment of the same product under different use scenarios. For the Mac product line, it distinguishes between fixed workstation office use and portable office use scenarios, adjusting the design parameters accordingly. According to Figure 7, which shows emotional design parameters and related user satisfaction attributes, it is evident that regardless of whether the device is portable, the product color significantly affects user satisfaction. The appearance of the design is always a key focus for users. In the global experimental environment, it is evident from the comparison between the MBP13 and MBA13 control groups, both with nearly the same screen size, that a reduction in product width and thickness can enhance user satisfaction with the product’s handling. Comparing the MBA13, which has the highest satisfaction level, with the MBP16, which has the largest screen size, reveals that the extra-large screen design of the MBP16 significantly increases its width, body thickness, and weight, yet does not decrease user satisfaction with the MBP16’s handling. The changes in user satisfaction relative to the product’s weight, thickness, and width are relatively minor compared to the impact of screen size design. An increase in product thickness due to a larger screen does not significantly affect user satisfaction with handling. This indicates that, for notebook product design, users prioritize screen size over the weight and thickness of the product. The design parameters related to weight and thickness are not particularly sensitive in terms of affecting user satisfaction.

5. Conclusions

5.1. Conclusions of the Study

This study of Apple’s product series serves as an illustrative example of the integration of emotional design and sustainable design principles. This integration was employed to develop a BP neural network design parameter optimization model that is based on emotional design and sustainable design. This model was designed to optimize the design scheme for designers for different products and materials. Subsequently, the design parameters were extracted in accordance with the principles of sustainable design and emotional design. This was followed by the administration of a questionnaire to ascertain users’ satisfaction with the various design parameters. Finally, the method’s efficacy was validated through the analysis of the mean square deviation (MSE) of the constructed BP neural network design parameter optimization model, as well as the discrepancy between the predicted and actual values of the model. The principal findings are as follows:

• The BP neural network is capable of analyzing the mapping relationship between emotional design parameters and sustainable design parameters for the various indicators of a product. Furthermore, it is able to accurately predict the optimal design solutions for products made of different materials under different usage environments.
• The extraction of design parameters for emotional design and sustainable design allows for a more targeted approach to product design than that based on a designer’s own design experience. The BP neural network optimization model, derived from the user’s satisfaction with various product indicators, enables the designer to effectively derive the user’s satisfaction with the emotional design parameters caused by the sustainable design parameters.
• The construction of the BP neural network design parameter optimization model enables the prediction of user satisfaction with different design schemes. This is achieved by taking into account the user’s different use scenarios and design material parameters, the user’s sensitivity to the product’s various design parameters and design parameter thresholds, and a reasonable range of optimization for the product’s design scheme.

5.2. Limitations of the Study

Although the BP neural network prediction model in this study has an accuracy of approximately 95% in predicting user satisfaction for design cases, this study primarily utilized Apple products for case studies, resulting in a relatively limited data source. At the level of product emotional design, Apple products are renowned for their simplicity and minimalism, which may reduce the emotional arousal of users. Conversely, a more aesthetically pleasing design in terms of appearance may lead to reduced user sensitivity to product size parameters. In terms of sustainable design, Apple utilizes aluminum alloy as a sustainable material. Furthermore, there is only a slight difference in the material properties between different generations of products using aluminum alloy. This experiment compares the design options of different generations of products under the parameters of aluminum alloy, as well as the use of glass back panels made of non-sustainable materials. This comparison may affect the universality of the conclusions.

5.3. Future Research Directions

In this experiment, the BP neural network prediction model based on emotional design and sustainable design achieved an accuracy rate of more than 95% in analyzing Apple’s product experiments. However, the data source of this experiment is relatively limited, introducing certain constraints. Concurrently, the design parameters extracted from this experiment based on the concepts of emotional design and sustainable design are relatively limited. In the future, it is necessary to achieve highly personalized product design predictions that cater to the unique needs and preferences of different user groups to meet the diversified market demand. In the process of product design, real-time data should be collected and fed back into the BP neural network model to enable the dynamic optimization of the design scheme. Additionally, efforts should be made to improve the interpretability of the BP neural network model, allowing designers and decision-makers to better understand the model’s decision-making basis and prediction results. This will facilitate more effective product design.