Low-carbon design based on conceptual design methods for mechanical and electrical products

This paper presents a conceptual product design method that considers the carbon emission factors of mechanical and electrical products. This method aims to consider low carbon issues prior to product design. Based on a comprehensive analysis of conventional product design procedures, the study extracts design information related to greenhouse gas (GHG) emissions, itemizes and quantifies the information, and ultimately uncovers the relationship among carbon emissions, function, and economic factors of newly designed products. This study explains the proposed concepts of carbon efficiency (CE) and carbon factor and combines them with the product design process to present a low-carbon design process of mechanical and electrical products. Furthermore, this paper presents a low-carbon design method of mechanical and electrical products, drawing on the product family design idea, with GHG efficiency and cost constraint. In particular, GHG efficiency pays attention to product function realization efficiency and economic factors. The potential value of the method is demonstrated in a case study of a tomato picking robot.


Introduction
Addressing climate change has become an urgent challenge facing human society in the twenty-first century. The 27th Conference of the Parties to the United Nations Framework Convention on Climate Change (COP27) [1] was held in 2022 to address this issue. The symposium aims to promote carbon peaking and carbon neutrality. In terms of carbon emission structure, China's carbon emissions [2] are mainly concentrated in industry (41%) and energy (46%). Therefore, industry is the main area of CO 2 emissions, and industrial low-carbon transformation is the key to achieving carbon peaking and carbon neutrality. Low-carbon product design is considered an effective and attractive approach to improve the carbon emission performance of mechanical and electrical products due to its considerable impact on the product lifecycle.
However, in industrial practice, studies on the identification of systematic methods to develop low-carbon products and manufacturing systems are limited. In addition, considering carbon emissions, product functions, and even economic factors at the beginning of the mechanical and electrical product design is difficult. Engineers often follow traditional principles and methods and evaluate the success of low-carbon product design through established metrics. A dedicated low carbon design approach is required to deliver repeatedly successful low-carbon designed products for the world. From this point of view, one of our current major concerns is to develop conceptual methodologies prior to product design to reduce greenhouse gas (GHG) emission throughout the entire product life cycle [3] (from cradle to grave). A life cycle assessment (LCA) tool was developed for mechanical and electrical products to enhance its operability in green design.
In recent years, scholars at home and abroad have carried out many research and applications in low-carbon products. For example, Deng [4] et al. quantified the carbon footprint analysis of machine tool components based on material flow, energy flow, and environmental emission flow and analyzed the impact of processing parameters on carbon emissions to optimize the processing scheme. Bacenett et al. [5] of the University of Milan used the LCA method to assess environmental impacts such as carbon emissions from farming using two similar tractors, but the study focused only on carbon emissions at the product use stage, and a large number of previous studies have proven that the carbon 1 3 emissions of mechanical and electrical products are mainly concentrated in the use stage. Zhang et al. [6] performed carbon footprint calculations for the connecting units, but only focused on the product structure aspect. Many of these studies show that the academic circles are paying more and more attention to the low-carbon aspects of mechanical and electrical products. Therefore, the research content of this paper is of great significance, which is the research on low-carbon conceptual design method of mechanical and electrical products.
The following content is the literature review on the research methods concerned in this paper. A typical low-carbon design method based on conceptual design provides at least two aspects: (1) low-carbon optimal design of mechanical and electrical products or decision-making of low-carbon solutions; (2) knowledge expression support for low-carbon product design. Thus far, considerable research efforts have been devoted to these aspects.
Specifically, with regard to the research on the first concept, some studies directly carry out the low-carbon design of a product with the goal of improving its carbon footprint. Ai et al. [7] proposed a low-carbon conceptual design method from the perspectives of technical system and human use, in which an improved process of requirement elicitation and analysis is implemented initially, and then the improvement strategies are proposed to establish a low-carbon function structure. He et al. [8] developed and analyzed the low-carbon robots in their study by the optimal design of the mechanism to reduce its carbon footprint through its life cycle. He et al. [9] also presented a carbon footprint model and a low-carbon conceptual design framework and measured the effects of design parameters on the estimation of product carbon footprint quantitatively. The low-carbon conceptual design framework proposed in this study pays more attention to the carbon footprint accounting of products and the sensitivity analysis of design parameters and less to the low-carbon optimization design guided by product design. Some studies focused on solving the contradiction between low-carbon design and functional satisfaction. Ren et al. [10] proposed an effective configuration methodology for low-carbon design, which has effectively solved contradictions with innovative design schemes. Liu et al. [11] proposed a leader-follower interactive decision-making mechanism for distributed collaborative product fulfillment of low-carbon product family planning (PFP) and manufacturer loading balancing (MLB) based on a Stackelberg game. In addition, some studies focused on product low-carbon design decision-making issues. He et al. [12] proposed a low-carbon design approach that uses the multigraph and color marking to solve the problem of one edge assigned with multiple weights and the constraint problem of vertex combination in the decision space of a lowcarbon product design.
In the second field, Wang et al. [13] proposed a new method of platform planning, considering cost and GHG emission of a product family simultaneously. Furthermore, an optimization method was applied to make a significant trade-off between cost and GHG emission to implement a feasible platform planning. While this paper partly refers to the method thinking put forward by them, and on this basis, this paper conducts relevant research on CE indicators that pay more attention to economic factors, which will be explained in detail in this paper. He et al. [14] proposed a feature-based integrated representation model with carbon footprint information about function, effect, and structure concept and presented an integrated principle solution model for low-carbon conceptual design. This study focuses on the knowledge representation and organization method of principle solutions. Guo et al. [15] proposed a smart knowledge deployment method. The study mapped function requirements with granular clustered knowledge into a matrix to match discretized knowledge and selected the derived candidate concept schemes in three steps: conflict-based primaries, configuration, and carbon footprint ranking. This study belongs to the study of knowledge representation method and pays less attention to economic factors. Wang et al. [16] proposed a comprehensive carbon footprint model for embodiment design based on macro-micro design features; it has complete design information and helps to apply the existing carbon footprint model to the implementation design stage. Ren et al. [17] presented an effective similarity determination model to support low-carbon product design, to deal with the challenge of acquiring knowledge of similar cases, considering similarity and adaptability. Zheng et al. [18] proposed a knowledge-based integrated product design framework to support low-carbon product development and to explore and navigate the integrative design space, considering low carbon and knowledge in a holistic perspective. Wang et al. [19] proposed a novel multiple criteria decision-making method for low-carbon product design to clarify the coupling relationship between low-carbon decision criteria and fuzzily express the low-carbon product design schemes. Dunmade et al. [20] developed environmental impact assessment methods in a multi-life cycle framework using the design concept of design for X (DFX) paradigms, such as design for modularity, cost, assemblability, manufacturability, detachability, maintainability, reusability, and remanufacturability. Although product design considers environmental and economic issues and applied throughout the product life cycle, it mainly focuses on industrial ecology, and belongs to macro research.
In summary, firstly, the current research on the conceptual design of low-carbon products either focuses on the quantitative evaluation of carbon footprint of existing mechanical and electrical products, and fails to guide the low-carbon optimization design in the initial stage of product design. Secondly, it either focuses on the knowledge expression at the beginning of mechanical and electrical product design, but can seldom guide the parametric low-carbon design of products. Thirdly, at the beginning of product design, there are few studies that combine low-carbon design with economic benefits.
This study provides an optimal design method of mechanical and electrical products, which comprehensively considers the carbon emissions, functions, and economic benefits at the beginning of the product design, and can provide parametric guidance for low-carbon design of mechanical and electrical products. This paper initially elaborates on the three concepts proposed, namely, CE indicators considering carbon emissions and economic benefits of the mechanical and electrical products, carbon factors, and low-carbon product family design, in which CE indicators can well combine carbon emission with economic factors. Then, design information related to GHG emissions is extracted on the basis of comprehensive analysis of conventional product design procedures, and the relationship among the three concepts is explored. As a result, the procedures and method for low-carbon product design are put forward. Finally, the method is verified by taking a typical agricultural machine-tomato picking robot as an example.

Method introduction
The new product design method integrates the CE concept, carbon factor concept, and product family design concept, thereby enabling it to factor in GHG emission, cost, and functionality. The carbon factor, which describes carbon footprint accounting, is the basis for the calculation of CE that links the environmental factor of the product's GHG emission and economic factors, such as work efficiency. As for the product family design idea applied in this study, we take the product function realization as the primary premise, and then determine the low-carbon product design scheme according to the carbon and cost factors.
The CE concept was proposed by Li [21] in accordance with the eco-efficiency concept proposed by the World Council for Sustainable Business Development (WCSBD) in 2000. This study explains the relationship between GHG emission and cost. CE is defined as the production target achieved by the emission of a certain amount of GHG from the product manufacturing system. The concept of CE focuses on the production goal, that is, the economic benefits brought by the product, and it also focuses on the efficiency of product function realization.
Furthermore, the functions of electromechanical products are mainly to realize the integral traveling function and local action function. Regarding the economic benefits, the economic value-added obtained through the functions of mechanical and electrical products also needs to be considered. In addition, most GHG emissions of the mechanical and electrical products are from the use stage, which has been explained in the Section 1. Therefore, the carbon factor is simplified as the calculation of GHG emission in the use stage (power consumption multiplied by carbon emission factor), which will be explained again in the concept of carbon factor subsequently. Therefore, considering the function realization, economic benefits of electromechanical products comprehensively, and carbon emission of mechanical and electrical products, this paper puts forward three CE indexes, namely, effective travel rate carbon efficiency (TRCE), effective action realization rate carbon efficiency (ARRCE), and economic return rate carbon efficiency (ERRCE), which can be expressed as follows: where TRCE is the TRCE, and it represents the increased travel rate per unit of carbon emissions.t is the travel rate of the product or equipment, which represents the overall traveling performance of the product.
P is the average power of the product. EF elc is the carbon emission factor of electricity, which is the intermediary that converts electrical energy into GHG emissions and varies by regions: where ARRCE is the ARRCE, defined as the product effective action realization rate and the average unit time carbon emissions of the product. It characterizes the speed at which the local action of the product is realized.a is the effective action realization rate: where ERRCE is the ERRCE, defined as the ratio of the economic profitability of function realization to the average carbon emissions per unit time of the product value.v is the economic return rate of mechanical and electrical products; it refers to the economic value-added realized by function realization in unit time. It is equal to the ratio of the added value given by realizing unit action to function realization time.
In addition, there are relationships among η TRCE , η ARRCE , and η ERRCE , where η ARRCE is equal to η TRCE multiplied by the action realization density, and η ERRCE is equal to η ARRCE multiplied by the value of unit action realization. The relationship among the three indicators is shown in Fig. 1.
Specifically, mechanical and electrical products in different fields have different functions. The selected laboratory has been studying the facility's agricultural machinery and equipment for more than 15 years. On this basis, this study focuses on the typical agricultural machinery equipment in the industry--tomato picking robot, which has great research value. The performance factors of product function efficiency are the coverage of fruit picking, the effective picking speed [22], and the economic return brought by picking. The three CE indicators designed for tomato picking robot are picking coverage carbon efficiency (PCCE), picking rate carbon efficiency (PRCE), and ERRCE.
Carbon factor is a significant concept based on ecological footprint [23] accounting method. It is defined as a coefficient mathematical model for calculating product GHG emission, which is measured in CO 2 equivalents (CO 2 e) during its entire life cycle. The life cycle-based carbon factor research is the basis of further studies.
The third concept is low carbon product family design, which is well developed and provides a logical path for product design. It has become a very important tool in our design method. Product family design uses the design idea of serialized products to carry out product development. It is a design method used to satisfy the requirements of different market segments. It aims to use the product platform to provide diversified products to the market, that is, a top-down design process.

Design procedures
The low-carbon product design method has five stages, which contain nine steps [24] (Fig. 2). In the specific implementation process of low-carbon design, various design methods, technologies, and tools are applied in different stages of the design process in a targeted manner.
The detailed procedures for designing a low-carbon product are as follows: Step 1. Product target selection. The product target selection includes selecting low-carbon design objects and defining the scope of object product design. The selection of product objects requires considering the technical feasibility, cost, and stakeholders of implementing a low-carbon improved design. This step corresponds to the system boundary determination step of the LCA, a common tool for product environmental impact analysis.
Step 2. Benchmark product determination. After determining the product target, a low-carbon reference product should be identified; it can be a virtual conceptual prod-uct. Its function is to provide a reference basis for evaluating the GHG emissions, economics, and technical level of product functions of new products. The determination of reference products is a key step in the comparison and demonstration between product systems because low-carbon products are a relative concept, and only relying on some absolute values cannot accurately evaluate the lowcarbon improvement results of the products. Therefore, a benchmark product needs to be determined to facilitate the subsequent design process.
Step 3. Low carbon demand analysis. Prior to the lowcarbon design, a detailed analysis of the status of the target product, such as market analysis, environmental protection laws and regulations, new technologies, and product information, should be conducted. These analyses help to understand user needs and provide direction for subsequent low-carbon designs.
Step 4. Inventory creation and analysis. From this step into the conceptual design stage, the product life cycle (including five stages: raw material acquisition → manufacturing → transportation → use → recycling), lowcarbon design factors (including resource utilization, energy efficiency, disassembly, and recycling), product categories for the establishment, and analysis of checklists can be comprehensively considered to determine the low-carbon design strategy (Step 6).
Step 5. Low-carbon design tools. The implementation of low-carbon design requires the support of corresponding design tools that can be applied to the entire life cycle process. Low-carbon design tools provide product designers with design guidelines to follow and easy-to-operate design tools to achieve more efficient and orderly lowcarbon designs. Commonly used low-carbon design tools include LCA tool, materials, energy, and toxicity (MET) matrix, and low-carbon design checklist.
Step 6. Low-carbon design strategy identification. This core content of low-carbon design is the process of translating the improvement requirements in the checklist into product design requirements. This step is mostly used in product family design.  Step 7. Low-carbon design plan formulation. This step corresponds to the preliminary design stage, which further concretizes the low-carbon design strategy into a low-carbon design scheme. This step requires the help of key technologies, such as material selection, structural design, disassembly, and recycling.
Step 8. Product detailed design. The detailed design of the product should be implemented into the specific implementation of the product design, the material selection of the product, the determination of the structure and its size, and the selection of the production process. Moreover, the feasibility of the design scheme is evaluated (Step 9).
Step 9. Design analysis and evaluation. In this step, a lowcarbon evaluation index system should be established, and the index weight should be determined. The low-carbon evaluation index system includes technical indicators, economic indicators, and carbon emission indicators.

Relationship of the three concepts
After completing the functional requirement analysis of the product and developing all the carbon factors to calculate the GHG emission, determining the relationship becomes simple. Function realization cannot be the only rule in designing a low-carbon product.
We should also maintain the relationship between GHG emission and cost. Figure 3 shows the relationship among the three concepts.
Therefore, through the product family design concept, a platform set matrix CP is utilized to describe the product family planning scenario as follows: where a, b, …, x indicates the number of optional instances of function modules {M1, M2, … Mk}.
The product family design provides some theories and methods to filter the function module configuration scenario to obtain the CPs. Thus, an optimized set of configure scenarios, represented by matrix D, can be obtained Through the multi-objective genetic algorithm, a functionally possible set matrix F is transformed into matrix D with cost and GHG emission constraints as follows: Matrix F is a set of configuration solutions that only considers the feasibility of functions, and then needs to be constrained by a certain proportion of cost and GHG emissions to obtain an optimized set of configuration solutions D.

Example: conceptual design of a tomato picking robot system
An example of tomato picking robot system illustrates the application of the approach to develop a common platform. Different from traditional tomato picking robots, which mostly use rigid joint structure, our research group has developed a flexible picking robot with high flexibility, low quality, and low energy consumption. We analyze the system-level functional requirements of the picking robot and obtain the subsystem requirements of the picking structure. Subsequently, the design of the walking mechanism, the lifting mechanism, and the dual picking robotic arms on the same platform were determined, and the construction of the overall structure was completed, as shown in Fig. 4. After the overall structure design, the seven modules of the tomato picking robot are determined, as shown in Fig. 5. On this basis, this study carries out the low carbon design of the product. Figure 6 shows each functional module candidate instance and each product variant initial module instance configuration.
The production quantity of product variants is shown in Table 1.
When each product variant in the product family is configured according to the initial module instance, the total cost calculation result of the product series is shown in Fig. 7, indicating fixed costs including design and development costs and variable costs, such as raw material costs, manufacturing costs, and assembly costs. The specific cost modeling and calculation process refer to the accounting methods provided by Santos et al. [25] and Fujita et al. [26], and the optimized calculation results are the same. According to the initial cost results in Fig. 7, the cost of the initial configuration of the product family is the largest because the sharing of the product modules is not considered in advance.
Its CE [27] is shown in Table 2. The data of η PCCE , η PRCE , and η ERRCE are calculated according to the actual picking speed, running power, the value of tomatoes picked before and after the optimal design of the tomato picking robot adopted in this case, and the electric carbon emission factor EF elc [29] in North China. The optimized calculation results are the same.
The following shows the design and product platform planning process for product families: In this study, the weights of CE and costs were set to w 1 = 0.5 and w 2 = 0.5, respectively. This focuses on the design process and methods that consider carbon emissions and costs in the low-carbon design of the mechanical and electrical products. Therefore, the research on the weight selection of CE and cost is still insufficient. In this study, the research is discussed according to a weight distribution situation set by Wang [30] in his research (w 1 = 0.5, w 2 = 0.5). Scholars in related fields can assign different weights to achieve deeper research in decision-making evaluation and other fields.
In CP jk , the larger k value indicates better performance of the module. Therefore, in a specific instance configuration, the value greater than or equal to its k value can be used as a functional candidate instance. Therefore, according to Fig. 6, the product variant F(3) is a feasible Fig. 4 The overall function structure of the tomato picking robot  Table 3. The same applies to other product variants.
Candidate design schemes of F(3) can be obtained by exhaustive method, as shown in Fig. 8.
Incompatible instance combinations were excluded to obtain the product planning platform F (Fig. 9).
The platform set matrix F and optimized product family planning scheme D are shown as follows:  6 Module instance configuration Table 1 The production quantity of the product variant Product variant Production quantity 4 5 3 2 4 8

Fig. 7
Cost distribution of product families by initial instance configuration Its CE (Table 4) and costs (Fig. 10) are as follows: This product solution considers functionality, CE, and costs. Compared with the initial example configuration, this solution enables the product design to perform better in terms of CE. The comparative analysis results of cost and CE before and after low-carbon optimization design are shown in Figs. 11 and 12, respectively. Figure 13 shows the final physical image of the product.

Conclusions
In this study, the three concepts proposed were described in detail, namely, CE, carbon factor, and low-carbon product family design. Among them, CE is an indicator that  Fig. 9 Obtaining the product planning platform F comprehensively considers carbon emissions and economic benefits. CE indicators were proposed according to mechanical and electrical product functions in different industries, which are TRCE, ARRCE, and ERRCE, and the correlation among them was illustrated. After the conceptual explanation, new low-carbon design procedures for mechanical and electrical products were proposed based on the traditional product design procedures, including five stages: requirement analysis, concept design, preliminary design, detailed design, and analysis and evaluation; they were also refined into nine implementation steps. Subsequently, this study focused on the second stage, namely, the concept design stage, where the relationship among the three concepts related to low-carbon design proposed above was explained. In addition, the matrix model of the low-carbon design of the mechanical and electrical products was proposed, including the initial product module matrix CP, for the matrix F that satisfies the compatibility requirements after functional constraints. Finally, the carbon emission and cost optimized matrix model was introduced for product configuration plan D. A typical agricultural machine, tomato picking robot, was used as an example to verify the method. In the concept design stage, the functional modules of the mechanical and electrical products were determined, from the low-carbon design was carried out. This paper presented the cost and CE of the initial configuration and optimized design of the tomato picking robot. The result shows that the total cost of the low-carbon design is not remarkably different before and after the design, whereas the CE η has increased by 31.12%, indicating that the low-carbon performance of the product has been significantly improved. Follow-up scholars can conduct more comprehensive research on optimizing cost and CE.
In summary, this study provides a conceptual design method for which the carbon emission, functionality, and economic factors of the mechanical and electrical products may be comprehensively considered at the beginning within the field of design. This approach could fill the gap of the low-carbon product design. The method was verified by taking a typical agricultural machine, tomato picking robot, as an example, and the parametric guidance of its design was realized.
Nomenclature GHG: Greenhouse gas; LCA: Life cycle assessment; DFX: Design for X; PFP: Product family planning; MLB: Manufacturer loading balancing; PKMs: Parallel kinematic machines; WCSBD: World Council for Sustainable Business Development; CE: Carbon efficiency; TRCE: Effective travel rate carbon efficiency; ARRCE: Effective action realization rate carbon efficiency; ERRCE: Economic return rate carbon efficiency; PCCE : Picking coverage carbon efficiency; PRCE: Picking rate carbon efficiency; CO 2 e: Carbon dioxide equivalents (kg ha −1 ); MET: Materials, energy, and toxicity Author contribution All the authors contributed to the study conception and design. Material preparation and analysis were performed by Guohua Gao, Xue Yang, and Zihua Zhang. The first draft of the manuscript was written by Xue Yang. A critical review, with modifications in the content and a thorough revision of the text of the manuscript, was performed by Zihua Zhang, Guohua Gao, and Zhenjiang Zhu, yielding the final submission version of the manuscript. All the authors read, revised, and approved the final manuscript.

Declarations
Ethics approval Not applicable.

Consent to participate
The authors declare their consent to participate.