Design Strategies for Modular Demountable Building Products Oriented to Design for Manufacturing and Assembly: A Case Study of M-Box1.0

Abstract

With the advancement of building industrialization and sustainable development, modular demountable buildings, as an efficient and environmentally friendly form, show significant potential in scenarios such as emergency housing and rural construction. However, they face issues including insufficient component adaptability, low demounting efficiency, and low integration level. Based on the Design for Manufacturing and Assembly (DFMA) theory, this paper proposes solutions and takes M-Box1.0 as a case study to explore design strategies from four dimensions: product modularization, logistics optimization, rationality of demounting, and component integration. The results show that M-Box1.0 has excellent ventilation and lighting performance. Compared with similar products on the market, it has fewer parts and lower costs. Moreover, it reduces construction waste through prefabrication and demountable connections. This study clarifies the advantages of DFMA-oriented design and has practical significance for promoting the efficient and energy-saving development of building industrialization.

1. Introduction

1.1. Research Background

As the “hardest hit” area in terms of energy consumption and carbon emissions, the construction industry is facing severe environmental pressures. Data shows that the industry consumes nearly 40% of the world’s available energy and produces one-third of the world’s total greenhouse gas emissions [1,2]. Of these emissions, more than 50% are from carbon dioxide emissions associated with buildings and raw materials [3,4]. Statistics from the International Energy Agency (IEA) reveal that, due to their large number and total floor area, residential buildings account for a much larger share of global energy consumption than commercial buildings do [5]. Against this backdrop, reducing the building sector’s energy demand and minimizing buildings’ environmental impact while improving the design process has become a critical issue that needs to be addressed. In the face of such challenges, modular demountable buildings are necessary. Their efficient, flexible, and environmentally friendly characteristics can accurately respond to the diverse needs of emergency housing and rural construction and provide an important solution to alleviate the construction industry’s environmental pressure [6].

The application of DFMA (Design for Manufacturing and Assembly) theory has given the construction industry a solid foundation for its transformation [7]. This design concept, which has a long history in the manufacturing field, can effectively solve the long-standing problems of low productivity, time delay, cost overruns, and poor safety in the construction field [8]. This paper takes the M-Box1.0 building product as a research case, focusing on the application of DFMA theory in modular demountable buildings. Through three steps—literature review on modular demountable buildings, strategy development, and effect verification—it reveals how DFMA theory achieves in-depth digital and industrial integration of building products from design to construction. The study aims to verify the practical value of DFMA theory in the construction field, explore the significant advantages of modular demountable buildings in terms of performance, cost, and sustainability, provide a reference technical paradigm for the development of similar building products, and promote the development of building industrialization towards higher efficiency and lower energy consumption.

1.2. Literature Review

1.2.1. Building Products Design for DFMA

The core of DFMA lies in systematically considering manufacturability and assembly convenience during the product design phase. By simplifying product and component structures, it enhances manufacturing and assembly efficiency while minimizing obstacles in manufacturing and assembly stages to the greatest extent possible [9]. Originating during World War II, this concept was first applied in weapons production by companies like Ford and Chrysler before evolving into a core principle of manufacturing design. Through forward-looking planning of manufacturing and assembly processes during the design phase, it achieves multiple objectives, including production optimization, cost reduction, and efficiency improvement [10].

With the advancement of construction industrialization, academia and industry have continuously extended DFMA from manufacturing to the construction sector. Cao [11] developed an ontology- and rule-based DFMA framework for wooden modular projects, demonstrating collaborative optimization pathways between design and manufacturing; Han Dongchen [12] established a systematic foundation incorporating DfM (Design for Manufacturing) and DfA (Design for Assembly), integrating IPD, parametric design, and BIM collaboration technologies; Chen Gang [13] applied DFMA to metro parking structure component design, enhancing prefabricated component performance through standardization and parametric modeling; The “full-process R&D + integrated team + information platform” model proposed by Cong Meng [14] aligns closely with DFMA’s core ideology through its life cycle collaborative logic.

The integration of prefabricated/modular buildings with DFMA highlights advantages over traditional construction methods. Studies show DFMA-optimized prefabricated construction improves convenience, shortens schedules, reduces labor demand, enables precise quality control, and minimizes material waste [15,16,17,18,19]. Modular construction transfers core components to factory-based standardized production, achieving full modularization or integrated construction with wide applications in education, commerce, healthcare, and other scenarios [20,21,22]. In practice, the kindergarten project at Guangzhou Baiyun Airport Phase III adopted the CMC new system, where 57 modules completed full-process production in factories, significantly improving on-site assembly precision [23]; CIMC’s modular buildings achieved over 90% prefabrication rate, shortened construction cycles by 50%, and reduced construction waste by 50% and carbon emissions by over 75% [24].

Prefabricated buildings demonstrate remarkable environmental and economic performance: structures with prefabricated walls reduce construction waste by 12.22% and energy consumption by 8.33% [25]; Cao’s [26] research showed prefabrication methods achieve 20.49% energy savings and 35.82% resource consumption reduction; a European study indicated prefabricated buildings cost 30% less than traditional structures over a 40-year life cycle [27].

While DFMA applications in construction have progressed from theory to practice, existing research primarily focuses on partial optimization during the design phase. Insufficient attention has been paid to its full participation in manufacturing, transportation, and assembly processes of modular detachable buildings, as well as proactive elimination of manufacturing–assembly conflicts. This gap provides a clear direction for research on relevant design strategies.

1.2.2. Modular Demountable Buildings

Modular demountable building is a deep integration of construction industrialization and the concept of sustainable development, which breaks down the building into reusable functional modules, and the core lies in the efficient use of the whole life cycle of the building through the standardized modules and demountable connection technology [28]. Demountable buildings can reduce the amount of construction waste generated through the recycling of materials, showing great potential for development, which has the characteristics of high flexibility, reusable, fast construction speed, and environmental protection and energy saving. Many scholars have carried out related research: Yan Hongliang [29] put forward the innovative concept of “N times removable mobile building is equivalent to permanent building”; Hu Feipeng [30] researched the construction technology of removable dry-hanging stone curtain wall and developed a convenient method of removable stone curtain wall construction. In the actual project, the design scheme of removable space steel structure of Vanke sales office in Guangzhou South Station [31] and so on, have accumulated experience in building composition, component connection, assembly process and so on.

However, currently demountable buildings are still facing high cost problems, for example, Galle’s [32] study showed that demountable building refurbishment methods cost 16% more than traditional methods, and the initial cost of the interior demountable wall type is 3% more, showing that the technology is not yet fully matured; FN Rasmussen [33] compared the environmental impact of recycled-materials buildings with demolition oriented design (DFD) buildings, and the study found that there is currently no environmental advantage to DFD buildings. Therefore, it is necessary to carry out systematic optimization throughout the whole life cycle of modular demountable buildings, from manufacturing to assembly, which is highly compatible with the mature DFMA theory in the manufacturing field. In addition, there is still a theoretical gap in the existing literature on specific DFMA strategies for modular demountable buildings. Therefore, DFMA strategies need to be constructed based on the characteristics of modular demountable building products in terms of construction and manufacturing.

Numerous studies have focused on the environmental benefits of demountable buildings. In the construction phase, Labik [34] noted that factory-based production significantly improves material efficiency. In practical projects, China Construction Science and Technology’s Shanghai project reduced on-site waste by 80% [35], Yantai Coastal Health Station cut construction waste by over 75% [36], and CIMC’s integrated system reduced construction waste by more than 90% [37]. Meanwhile, Giovanni [38] proposed that demountable buildings are more efficient to construct, enabling shorter construction periods. Through comparative verification, Yang [39] found that standardized processes not only improve efficiency and shorten construction periods but also reduce labor and management costs. At the end of the full life cycle, Lisa’s [40] research showed that 70–90% of materials in demountable buildings are easy to disassemble and recycle at the end of their design life, significantly reducing waste pressure and lowering raw material demand. BuildGreen’s long-term research also confirmed high recyclable material ratios and convenient recycling during demolition, effectively promoting resource circulation [41]. The International Institute for Sustainable Construction reported rising global application of such buildings, with notable advantages in resource conservation and environmental impact reduction—for example, a BoKlok residential project cut full-life cycle carbon emissions by ~35% vs. traditional buildings [42]. Zhang [43] found demountable buildings quickly meet desert emergency needs and reduce air conditioning energy use, showing strong adaptability and sustainability. In summary, demountable buildings reduce consumption and improve efficiency across the life cycle, demonstrating multiple environmental sustainability values.

1.3. Research Purpose and Structure

This study aims to explore the design strategy of modular demountable building products based on the DFMA theory, and to verify it in practice through the typical case of M-Box1.0. Specifically, the first is to construct a systematic DFMA design strategy framework, covering four key dimensions: modularization of product systems, rationalization of assembly methods, optimization of logistics and transportation, and integrated design of components; the second is to demonstrate how these strategies are transformed into actual architectural product design through the design practice of M-Box1.0, while the third is to analyze the effects of M-Box1.0 in terms of performance enhancement, shortened construction period, cost reduction, and sustainability enhancement, and to provide an opportunity for the development of a new design strategy for building products. The study also quantitatively analyzes the effect of M-Box1.0 in terms of performance improvement, construction period reduction, cost reduction, and sustainability enhancement, etc., to provide a scientific basis and practical examples for the design optimization of modular dismantlable building products, and help the industrialization of construction develop in the direction of high efficiency, energy saving, and environmental protection.

This paper is divided into five main parts. The first part is the introduction, which explains the importance of modular demountable buildings in the context of construction industrialization, combs through the current status of the application of DFMA theory and research gaps, and introduces the research purpose of this paper. The second part introduces the research methodology, detailing the four design strategies proposed based on DFMA theory and their theoretical basis and application logic. The third part focuses on the design practice of M-Box1.0 based on DFMA, combining with practical cases to explain the specific implementation and innovation of each strategy in product design, with relevant charts to assist understanding. The fourth part verifies the implementation effect of M-Box1.0, quantitatively analyzing the four aspects of performance, duration, cost, and sustainability, and highlighting its advantages compared with traditional buildings. The fifth part is the conclusion, which summarizes the research results, clarifies the performance enhancement achieved by M-Box1.0 through the DFMA strategy and its value to the development of the industry, and at the same time, points out the limitations of the research and the future research direction, which will provide a reference for subsequent research.

2. Methods

Aiming at the shortcomings of the current application of DFMA theory in the field of modular demountable buildings, this study combines the core idea of DFMA with the construction characteristics of building products, and puts forward four targeted design strategies, which aim to optimize the whole process of modular demountable buildings from factory manufacturing, logistics and transportation to on-site assembly, so as to achieve a balance between manufacturability, assemblability, efficiency, and sustainability (Figure 1).

2.1. Product System Modularization

This strategy simplifies the complex structure of building products by systematically dividing them into modules. The building is demountable into standardized and functionally clear modules to reduce component redundancy and clarify the logical connection between parts. The modular system adopts a hierarchical structure to adapt to the spatial combination of diverse scenarios (e.g., emergency housing and rural construction). This division not only facilitates the parallel production of multiple components in the factory but also simplifies the on-site assembly process workers can quickly assemble them according to a clear labeling and classification system; at the same time, the modular design supports users to customize the spatial configuration according to their needs, which strikes a balance between standardized production and personalized needs.

2.2. Rationalization of Assembly Methods

Given that there is no absolute positive correlation between high prefabrication rate and easy assemblability, this strategy emphasizes balancing manufacturability and assemblability through technical optimization. Specifically, it includes the following aspects: it uses embedded angle brackets + riveted bolts for connection and pre-installs standardized hoisting rings, simplifying operations and adapting to small cranes; Relying on NASTRAN 2026 software, it meets the requirement of 100 assembly cycles, with a safety factor of more than two for key parts, complying with relevant specifications; it also uses parametric software such as Inventor 2026 to build detailed models by discipline, perform virtual assembly to adjust parameters, accurately control design errors, ensure component size matching, and reduce on-site adjustments and rework.

2.3. Optimized Design of Logistics and Transportation

Unlike manufacturing products, building components involve off-site production and on-site assembly, making transportation a key bottleneck. The strategy improves transportation efficiency by optimizing the size and packaging of the components—adopting the “split-piece transportation” mode (instead of whole box transportation), disassembling the modules into flat and regular components to maximize the loading capacity of a single vehicle and reduce the number of transportation times. The design of the components also prioritizes the use of simple and regular forms to reduce the number of corners and sharp structures, thus reducing the risk of damage during transportation and further controlling the cost. At the same time, the design of the components prioritizes the use of simple and regular forms, reducing corners and sharp structures, reducing the risk of damage during transportation, and realizing “near-zero-damage” delivery to further control costs.

2.4. Component Integration

Component-integrated design is grounded in material properties and structural logic, with the core of integrating multiple functions into a single component to enhance manufacturability and assemblability. Specifically, it reduces the number of components and on-site installation steps through function integration. For example, the “exoskeleton” structure of M-Box1.0 integrates decorative, structural, and waterproof functions. This design offers significant advantages: it boosts assembly efficiency, reduces high-altitude operations to lower safety risks, and improves the utilization of on-site equipment and tools, further simplifying the construction process.

The four strategies form a closed loop: “modularization defines the framework → rationalized assembly ensures delivery → optimized logistics breaks bottlenecks → integrated component design boosts full-process efficiency”. Specifically, modular design establishes the foundational “skeleton”; rationalized assembly methods serve as the “bridge” linking design and construction; optimized logistics transportation resolves cross-regional distribution pain points; and integrated component design realizes efficiency gains across the entire process.

3. DFMA-Oriented Design Practice of M-Box1.0

M-Box1.0 is a modular demountable building product based on the DFMA (Design for Manufacturing and Assembly) concept (Figure 2). In its name, “M” has a double meaning: it points to ‘module’, interpreting the modularity concept upheld by architectural design; and it is also associated with “magic”, highlighting the flexibility and convenience shown in the building’s assembly process—i.e., the ‘magic’ property of its disassembly. The “1.0” identifies it as the first-generation product.

M-Box1.0 has a wide range of product adaptations that can be applied to flexible space needs in B&Bs, residences, commercial activities, and specialized environments. The existing completed sample is equipped with many features for its application in the residential sector, including a living room, two bedrooms, a bathroom, and two outdoor terraces (Figure 3). With a building area of 59.5 m², the house occupies 45 m² and can be assembled in 2 days.

3.1. Modular Design of the Product System

The M-Box1.0 uses standardized modules and demountable connection technology (Figure 4). This technology overcomes the limitations of monolithic modular building products regarding space, transportation, and construction sites. The internal space and external shape can be adjusted according to actual needs, transforming from a single fixed mode to diverse variable modes. The modular components of the M-Box1.0 can be demountabled for transportation, effectively reducing the required road conditions and enabling the product to traverse complex environments, such as narrow, height- and width-restricted roads. The modular design of the M-Box1.0 enables assembly operations on smaller sites with complicated terrain, greatly expanding the application scenarios and promotional scope of modular, demountable building products.

Considering the dismountable nature of building products, even if the building scale is not large, the number of components after disassembly is still quite large. The complex connections between each component will inevitably lead to an increase in connection problems, which in turn affects the efficiency of manufacturing and assembly [23]. Therefore, designing a simple and clearly divided system is beneficial to improving the efficiency of both manufacturing in the factory prefabrication stage and on-site workers’ assembly. Such a design enables multiple types of work to be manufactured simultaneously, shortening the project cycle; the clear system classification allows workers to assemble according to the labels during assembly, and users can select configurations by category when purchasing. In response to this, this product constructs a three-level classification architecture for modular design (Figure 5): the first-level classification includes structural modules, appearance modules, equipment modules, and space modules; the second-level classification is further refined. For example, structural modules cover main structures (roof panels, wall panels, floor panels, internal partition walls) and auxiliary structures (foundations); appearance modules are subdivided into balcony, step, window, and handrail modules; equipment modules integrate air conditioning, photovoltaic, and wind turbine modules to meet the energy supply of the building; space modules include living room, bedrooms of different area specifications (9 m², 13.5 m²), dining kitchen, bathroom, and staircase modules. The third-level classification focuses on specific components, including both basic components, such as roof panels in the main structure and space components of different area specifications in the bedroom module, forming a hierarchical system architecture of “module-component-part”.

Through standardized connection technologies and clear hierarchical classification, multi-trade parallel operations are enabled at the manufacturing stage, while workers can quickly match and assemble components by category during installation. This logically reduces connection confusion, echoing the design consideration of the correlation between system functions and manufacturing efficiency. Meanwhile, relying on the clear “module-component-member” hierarchy, configurations of similar modules can be interchanged, and demountable components can be reused through standardized connections. This builds a complete detachable system featuring “standardized connections + module interchangeability,” reducing efficiency disruptions caused by connection issues and ensuring smooth manufacturing and assembly processes.

In the design architecture of M-Box1.0, each type of module has specific functions, and different configurations of modules of the same type can be interchanged. The selection of module sizes takes into account the ergonomic user experience and the specifications of road transportation. With 1.5 m and 3 m as the basic modules, a standardized splicing system has been established. At present, architectural design has transformed from the “architectural work design mode” to the “architectural product design mode” [44]. The design concept of M-Box1.0 is rooted in the philosophy of product design. Designers are responsible for constructing the logical framework of module combination, while owners can flexibly select and combine modules within this framework according to their specific functional needs to create personalized spatial effects, thus highlighting the close and effective collaborative relationship between design and manufacturing.

3.2. Rationalized Design of Assembly Methods

3.2.1. Optimization of Connection Nodes and Hoisting Design for Assembly Convenience

The design of the M-Box1.0 product, which connects two regular enclosure panels through embedded corner codes and blind rivet bolts (Figure 6), fully considers the convenience of assembly and installation. During installation, the top plate is pressed over the side plates, and the side plates rest on the bottom plate. The connection between them is achieved through the insertion of standardized L-shaped connectors, with the insertion position at the end of the main vertical ribs. They are fastened by high-strength bolts, and the position for embedding the waterproof rubber strip on the bent surface has been reserved through design. When the ends of the module openings are used to connect two modules, high-strength bolts are directly used to fasten the curled parts formed by the bending process. When modules are connected left and right, the side wall panel of one module is opened, and they are connected to each other with high-strength bolts. When modules are connected up and down, the top plate of the lower module is used as the bottom plate of the upper module. An opening is made at the upper part of the side wall rib, and a T-shaped connector is inserted and fastened with high-strength bolts. Through the combined application of these different connection methods, modules can be spliced into architectural forms with various shapes. Lifting rings are installed at the four corners of the top plate of each module, which match the hooks of the crane. The weight of each module panel is lower than the lifting capacity of small cranes, enabling quick and accurate transfer and lifting from the flatbed truck to the site (Figure 7).

Traditional connections require repeated alignment of bolt holes or fixing of welding positions, involving multiple measurements and calibrations. In contrast, embedded angle code connections achieve “plug-and-position” through pre-set insertion points (such as at the ends of vertical ribs or openings inside wall ribs), eliminating cumbersome alignment steps. While traditional connections demand step-by-step installation of various components or multiple welding procedures, this system uses only uniform high-strength bolts with standardized connectors. With significantly fewer component types, it can be directly tightened after insertion. Additionally, modules form a self-supporting structure via insertion, requiring no temporary bracing. Pre-reserved slots for waterproof strips enable simultaneous sealing during tightening, integrating multiple processes to enhance assembly efficiency.

3.2.2. Fatigue Analysis of Demountable Parts to Ensure Structural Durability

The durability simulation of the demountable parts of M-Box1.0 is carried out based on NASTRAN. The fatigue analysis adopts the stress-life method, with Von Mises stress as the evaluation index and a threshold of 100. Combined with the S–N curve, the ultimate strength of the material $S_u$ = 200 (in accordance with GB/T 228.1-2021 Metallic Materials-Tensile Testing-Part 1: Method of Test at Room Temperature [45]); the cycle base $N_0$ = 1000 and other parameters, as well as the load scale factor–design–construction time curve, are used to simulate the load changes during the repeated assembly process (Figure 8).

Finite element results (Figure 9) show the fatigue life cloud maps of L-shaped, T-shaped, and +-shaped splicing nodes. Except for local areas such as the edges of bolt holes and key stress-bearing parts, the fatigue life of most areas is greater than $1\times 10^{10}$ cycles, and the fatigue life of most other areas is also at a high level. With reference to the relevant guidelines on the number of repeated demounting and assembly of modular buildings in GB/T 51232-2016 Technical Standard for Fabricated Steel Structure Buildings [46] and combined with the actual application scenarios of the project, the number of repeated assembly cycles is determined to be 100. The yield strength of the material is in accordance with GB/T 700-2017 Carbon Structural Steels [47]; the yield strength of Q345 steel is 345 MPa. The maximum stress of key stress-bearing parts in the simulation is much lower than this value, and the safety factor of the parts is greater than two, indicating that the stress level meets the fatigue durability requirements. Therefore, the durable performance of the demountable parts of M-Box1.0 can ensure reliable use within the designed cycle of 100 times.

3.2.3. Inventor Modeling for Accurate Control of Design Errors

During the R&D phase of the M-Box1.0 product, the team relied on BIM technology and introduced Inventor software, commonly used in the manufacturing industry, to build models step by step. It conducted an integrated design of the building structure, enclosure system, various pipelines for water supply and drainage, heating, ventilation, and air conditioning (HVAC), as well as prefabricated interior furniture, panels, doors, and windows. By accurately setting parameters such as the size, shape, and material of each part (Figure 10), the basic modeling is completed. Then, the software’s part assembly function is used for virtual assembly of components, and the specific size and design of components are adjusted in real time according to the matching accuracy between components. For example, when a large gap is found between the wall panel and the roof panel, the position parameters of the reinforcement at the top of the wall panel are adjusted immediately, and the connection method is optimized to eliminate design defects in the virtual phase, ultimately generating a complete set of construction drawings. The manufacturer directly manufactures components according to the drawings, and the construction and assembly party directly guides the construction based on the drawings, ensuring a close connection between all links and effectively controlling design errors accurately.

3.3. Logistics and Transportation Optimization

3.3.1. Design of Disassembled Sheet Transportation to Improve Loading Efficiency

The M-Box1.0 fully considers the efficiency of logistics and transportation during the design phase. Most modular building products adopt integral box transportation, which often faces difficulties in passing through narrow mountain roads and complex road conditions. In contrast, the M-Box1.0 disassembles building modules into panels for transportation (Figure 11), improving assembly efficiency and the recyclability of components. Compared with building products using integral box transportation (one box per vehicle), this design increases the load capacity of a single vehicle by nine times and reduces transportation costs by 60%. All components of the 59.5 m² sample house can be transported in a single trip by one vehicle. To further ensure transportation convenience and “near-zero damage” of the product, the M-Box1.0 follows the principle of “simplicity and regularity” in component shape design: based on the module division logic, each main component is ensured to have a regular shape, minimizing corners and sharp structures to reduce the risk of collision damage during transportation and facilitate stacking and placement to maximize the use of transportation space.

3.3.2. Integrated Design of Plumbing, Electrical, and HVAC

In response to the demountable nature of the building, targeted adjustments were made to the water supply and drainage, electricity, and HVAC systems during the design process. In areas with dense pipelines, such as bathrooms, the integral box transportation method is adopted to realize complete factory prefabrication; water, electricity, and HVAC pipelines in other spaces are integrated into the enclosure structure. By presetting standardized interfaces and junction boxes with protective covers at both ends of the wall, a “plug-and-play” connection is achieved during on-site enclosure panel splicing, significantly reducing the cross-operation of on-site pipeline layout.

During the design phase, the parametric modeling function of Tiangong DFC 5.0 software was used for core design work. Through this software, the collaborative optimization of pipelines and wall structures was realized, and the path layout of water, electricity, and heating pipelines was optimized accurately. This not only significantly reduced the total length of pipelines but also minimized the number of times water and electricity pipelines pass through multiple box-type rooms, ensuring the accuracy of interface positioning and integration feasibility. For the air conditioning system, the design adopts an independent module and selects an integrated air conditioning unit to realize integrated air inlet and outlet, avoiding interference with the design of other professional systems. The centralized layout logic of water, electricity, heating, and ventilation pipelines is displayed through the “DFC Centralized Layout Diagram of Water, Electricity, and Heating Pipelines” (Figure 12). In the diagram, green and purple pipelines correspond to cold and hot water pipes; red pipelines are electrical wires, and blue and yellow pipelines represent floor heating pipes. To highlight the centralized layout logic, the local pipeline routes inside the walls in the diagram are simplified.

After completing the integrated design of water, electricity, heating, and ventilation pipelines, collision detection was carried out based on Tiangong DFC software (Figure 13). The number of collision points was reduced from 16 to 0 after optimization, and a distribution table of collision types is attached. The detection report shows that all collision points in the building have been resolved after optimization, and there are no existing collision problems.

3.4. Integrated Design of Components

Traditional building walls are generally constructed using a layered construction system, which requires the installation of independent structures such as a finish layer, structural layer, waterproof layer, thermal insulation layer, isolation layer, and interior layer, etc. The material connection and process coordination between layers not only increase the construction complexity, but also easily lead to leakage, thermal bridges, etc., due to improper interface treatment. M-Box1.0 introduces mature sheet metal technology in the automobile manufacturing field into the construction process (Figure 14). M-Box1.0 has developed an exoskeleton structure system, which realizes a high degree of integration of component functions through the in-depth integration of material characteristics and construction logic.

3.4.1. Design of “Exoskeleton” Structure for Function Integration

The “exoskeleton” structure of M-Box1.0 takes the thin-walled light steel structure as the core and consists of a set of steel plates with bent edges, ring-shaped frames, and reinforcing ribs. Its geometric characteristics are derived from the modular cylinder design: the rectangular frames of the top surface, side surfaces, and bottom surface form a closed cylinder through surrounding connections. Through the “rib retreat” design (shifting the columns and load-bearing lintels at the edges of traditional modules inward), only narrow bent surface structures are retained at the module connection parts. This design not only avoids material redundancy of “double beams and double columns” in traditional module connections but also naturally forms a space for screwdriver operation and a groove for waterproof rubber strips.

The ring-shaped frame, as a key component for structural reinforcement, is formed by integral bending of Q345 steel (in line with GB standards, with an elastic modulus of 2.06 × 10⁸ kN/m², a Poisson’s ratio of 0.3, and a bulk density of 76.98 kN/m³), and its cross-section is designed as a box shape of 100 mm × 50 mm × 1.2 mm (height × width × thickness). A three-dimensional model was established using Midas Gen 2024v2.1 software to verify its mechanical performance: under the combined working conditions of applying 1.3 times the permanent load (including the self-weight of the roof photovoltaic array of 0.25 kN/m² and the equipment point load of 1.5 kN) and 1.5 times the live load (including the snow load of 0.55 kN/m² and the roof live load of 2.0 kN/m²), the stress distribution range of the frame is −0.038~0.046 N/m², and the maximum deformation is only 0.053 mm, which is far lower than the limit of 1/250 of the beam span (12 mm) specified in GB50009-2012 [48], verifying the mechanical efficiency of the integrally bent structure (Figure 15).

3.4.2. Performance Verification of the “Exoskeleton” Structure

The finite element analysis of the stress on the “exoskeleton” structure shows (Figure 12) that the “exoskeleton” structure wall formed by bending 1.5 mm thick steel plates (the steel material used is Q345, in line with Chinese national standard GB/T 700-2017, with an elastic modulus of 2.0600 × 10⁸ kN/m², a Poisson’s ratio of 0.3, a linear expansion coefficient of 1/1.2000 × 10⁻⁵, a bulk density of 76.98 kN/m³, a mass density of 7.85 kN/m³/g, a thermal characteristic value of specific heat of 60 kJ/(kN·°C), and a damping ratio of 0.02) has excellent resistance performance.

According to GB50009-2012 Code for Loads on Building Structures [48], the standard value of uniformly distributed live load on the floor of residential buildings in civil buildings is not less than 2.0 kN/m². Therefore, when checking the longitudinal compressive strength of the building “exoskeleton”, a permanent load of 2.0 kN/m² was considered in addition to the gravity load. Finite element analysis using Inventor software shows that the maximum longitudinal deformation of the “exoskeleton” formed by bending 1.5 mm galvanized steel plates is approximately 0.657 mm (Figure 16a), which meets the qualification standard for floor static load tests in GB/T 50344-2019 Standard for Inspection Technology of Building Structures [49], i.e., “maximum deflection ≤ L/250 (L is the floor span)”. When checking the transverse compressive strength of the building “exoskeleton”, horizontal loads such as wind load and seismic load, as well as vertical loads such as self-weight, were comprehensively considered. According to the relevant provisions of GB50009-2012 Code for Loads on Building Structures, a transverse pressure of 0.025 MPa was determined. The maximum transverse deformation of the “exoskeleton” is approximately 10.78 mm (Figure 16b), which meets the specification requirements that for general profiled steel plate exterior walls, the maximum deflection (a form of deformation) under transverse pressure, such as wind load, should not exceed 1/150 to 1/200 of its span.

In addition, Simsolid 2025 software was used for visual analysis of the structural performance of modules connected between upper and lower layers through “T-shaped connectors + high-strength bolts”. The self-weight of the box body, the weight of the stacked box bodies, and an additional permanent load of 2.5 kN/m² were considered in the analysis. According to GB50009-2012 Code for Loads on Building Structures, the snow load in the 50-year recurrence interval in North China is 0.55 kN/m². Considering the possibility of long-term snow accumulation on the roof and the presence of people on the roof after vertical connection, the live load on the roof was considered to be 2.0 kN/m². The steel material used is Q345. After defining the load combination and conducting the module operation analysis, it was observed in the results that the ring-shaped frame system connecting the upper and lower layers is under uniform stress, and the maximum deformation of the module in the XYZ directions is only 1.1091 × 10¹ mm (Figure 17), which is far less than the deformation limit of 1/250 of the beam span, indicating that the expansion of the second floor between modules is feasible.

4. Discussion

4.1. Performance Analysis

The geometry of the barrel-type structure has windows and doors that open primarily at the ends. First, the non-load-bearing structure at the ends of the cylindrical structure allows for the opening of a large area of windows and doors, permitting sufficient natural light to enter, ensuring that the interior space is adequately illuminated, while preventing glare. In addition, the form of open windows and doors at the ends can utilize the more narrow and long interior spaces to create temperature differences, which facilitates natural ventilation throughout the building and enhances convective airflow (0.8–6 m/s). Through the ventilation and natural light analysis of M-Box1.0, the building product meets the standards in key performance indicators. The regular space shape of M-Box1.0 further enhances the wind pressure ventilation effect, which is confirmed by the results of the building wind simulation (Figure 18a). The building interior is able to form good natural ventilation paths when the doors and windows are open. Sufficient natural light enters to ensure that the interior spaces are well-lit while preventing glare. The natural lighting simulation shows (Figure 18b) that more than 90% of the spaces are illuminated at an average daily illuminance level of more than 300 lux. The lighting coefficients of M-Box1.0 all meet the requirements of the Building Lighting Design Standard (GB50033-2019) [50].

4.2. Analysis of Module Part Count

The “exoskeleton” structure of M-Box1.0 is an optimization based on the existing modular steel structure wall panels available on the market [51,52,53]. In comparison, the “exoskeleton” structure of M-Box1.0 uses full welding technology to replace most parts. Taking the splicing of a 3 m × 3 m wall panel and floor panel as an example (Figure 19), the “exoskeleton” structure of M-Box1.0 only requires 3 × 4 = 12 bolts of the same type to complete the assembly. In contrast, a modular steel structure wall of the same size requires 6 types of accessories, with at least 3 pieces of each type—meaning a minimum of 18 accessories to finish the assembly. It is evident that with the “exoskeleton” structure of M-Box1.0, the number of parts needed for assembling a 3 m × 3 m wall panel and floor panel is reduced by at least six.

By reviewing relevant literature and patents [51,52,53], it is summarized that a single existing modular steel structure wall panel module on the market contains 173 parts (

Table 1. Comparison of part counts between existing modular steel structure wall panel modules and M-Box1.0 wall panel modules.

Existing Modular Steel Structure Wall Panel Module (Single Unit)			M-Box1.0 Wall Panel Module (Single Unit)
Detailed Category	Specific Component	Quantity (Piece)	Detailed Category	Specific Component	Quantity (Piece)
Structural Components	Longitudinal Main Rib	6	Structural Components	“Exoskeleton” Plate	1
	Transverse Main Rib	2		Longitudinal Main Rib	3
	Transverse Secondary Rib	2		Transverse Main Rib	10
	Square Steel Gasket	8		Transverse Secondary Rib	10
	Vertical Steel Column	2
Connecting Components	Angle Steel	16	Connecting Components	Angle Bracket	3
	C-shaped Steel	16
	Uplift Component	4
	Cross Steel Strip	4
	U-shaped Steel	3
Bolts	Self-tapping Screw	64	Bolts	Shear Bolt	12
	Dovetail Screw	24
	Shear Bolt	32

), while a single M-Box1.0 wall panel module contains only 39 parts. It is clear that the M-Box1.0 wall panel module has fewer total components and fewer types of components. Assuming the standard assembly time for each component is the same, the assembly efficiency of M-Box1.0 is significantly higher than that of traditional modular steel structures.

4.3. Cost Analysis

The cost of building products mainly consists of material costs and labor costs. Data shows that the labor cost of manufacturing workers is only 35–65% of that of on-site construction workers [54]. Based on this wage difference, M-Box1.0 significantly reduces labor costs by transferring most of the manufacturing and assembly processes of building components to the factory. At the same time, the efficiency of the standardized assembly process in the factory reduces the idle and waiting time commonly seen in on-site construction, further reducing on-site assembly costs. In terms of material costs, the basic material costs for on-site construction and prefabricated installation are roughly the same. However, through the design of the “exoskeleton” structure, M-Box1.0 achieves a significant reduction in steel usage while ensuring unchanged structural strength, thereby making the overall material cost lower than that of traditional solutions. Nevertheless, DFMA-oriented prefabricated construction incurs additional costs such as design optimization and module packaging, which need to be allocated to the total building cost.

The M-Box1.0 building product is benchmarked against modular structural system building products with steel frame exterior enclosure and load-bearing features, launched by small- and medium-sized modular building enterprises on the market, including Liulangcang, Weisu, Xingjian, Huji, and Yincang. Through research on data from their official websites, the quoted price of Liulangcang [55] products is 6500 yuan/m²; the quoted prices of Weisu [56] and Huji [57] products are 6200 yuan/m²; the quoted prices of Xingjian [58] and Yinneng [59] products are 6800 yuan/m². The profit margin of steel structure projects is generally between 10% and 20% [60], and 15% is adopted for calculation. Excluding the profit, it can be concluded that building a 59.5 m² modular steel structure building requires approximately 329,000 yuan.

$$\mathrm{F}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{u}\mathrm{l}\mathrm{a} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{U}\mathrm{n}\mathrm{i}\mathrm{t} \mathrm{P}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{S}\mathrm{i}\mathrm{m}\mathrm{i}\mathrm{l}\mathrm{a}\mathrm{r} \mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{s} \mathrm{o}\mathrm{n} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{M}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t}={\displaystyle \frac{\left(6500+6200\times 2+6800\times 2\right)}{5}}(1-15\%)\approx 32.9$$

Combined with the actual construction records of M-Box1.0 (

Table 2. Construction Cost of M-Box1.0.

Prices and Classification		M-Box1.0
material cost	1-Structure	10 w (“exoskeleton” structure)
	2-Interior finishes	1 w
	3-Exterior finishes	0.2 w (painting/filming)
	4-Insulation	1 w
	5-Plumbing	1 w
	6-Kitchen and bathroom fixtures	1.5 w
	7-Photovoltaics and energy storage	2 w
labor cost	8-Fabrication	3 w
	9-Packing	0.1 w
	10-Transportation	0.1 w
	11-Assembly/building costs	0.2 w
add up the total		20.1 w

(Data is from on-site assembly video timing and worker operation records).

), its total construction cost is 201,000 yuan, which is only 61.1% of that of a modular steel structure building of the same size.

4.4. Sustainability

The DFMA-oriented design of the M-Box1.0 modular demountable building product takes into account the recycling and reconstruction of building materials from the very beginning. It integrates the building’s geometric shape, structural stiffness, construction method, and building operation energy consumption. By quantitatively calculating carbon emissions and resource recycling efficiency, it minimizes the energy consumption of the building throughout its life cycle. The structural design adopts an “exoskeleton” structure, achieving integrated design of decoration-structure-waterproof functions, which reduces material redundancy and construction complexity. The cross-scenario interchange mechanism of the standardized module units of M-Box1.0 supports the functional iteration and adaptive update of the building during its service life. This avoids waste caused by demolition and reconstruction due to functional changes and reduces material resource waste resulting from repeated decoration and unnecessary construction.

This paper uses the “Life Cycle Assessment Method” to calculate the carbon emissions of M-Box1.0 during the material preparation, component manufacturing, transportation, assembly, and operation stages. The key parameters and formulas are as follows (

Table 3. Carbon emissions during the construction and building operation phases.

Carbon emissions during construction	Material preparation (t.CO2)	16.2
	Component production (t.CO2)	3.44
	Transportation (t.CO2)	5.08
	Component assembly (t.CO2)	1.39
Carbon emissions during operation	Clean energy (t.CO2·y)	−32.11
	Building system energy consumption (t.CO2·y)	3.09

): the steel carbon emission factor involved in the formulas is derived from Standard for Calculation of Building Carbon Emissions (GB/T 51366-2019) [61]; the transportation carbon emission factor is from Guidelines for Calculating Carbon Dioxide Emission Intensity of Operating Vehicles in the Highway and Waterway Industry [62]; the power grid emission factor is from 2023 Regional Power Grid Baseline Emission Factors for CCER Projects [63]:

1.

The carbon emission formula for material preparation is:

E₁ = ∑(M_i × EF_i) = 6 t × 2.7 t.CO₂/t = 16.2 t.CO₂

(1)

where ${\mathrm{M}}_{\mathrm{i}}$ is the steel consumption, with a value of 6 t; $\mathrm{E}{\mathrm{F}}_{\mathrm{i}}$ is the carbon emission factor of steel, with a value of 2.7 t.CO₂/t.

2.

The carbon emission formula for component production is:

E₂ = ∑(Q_j × EF_j) = 6 × 0.573 t.CO₂/t ≈ 3.44 t.CO₂

(2)

where ${\mathrm{Q}}_{\mathrm{j}}$ is the component output of this project, with one module as a group and a value of 6 (M-Box1.0 consists of 6 modules); $\mathrm{E}{\mathrm{F}}_{\mathrm{j}}$ is the carbon emission factor of the component production stage (including comprehensive carbon emissions from raw materials, production processes, energy consumption, etc., combined with component types such as steel structure components and corresponding industry data).

3.

The carbon emission formula for transportation is:

E₃ = ∑(D_k × W_k × EF_k) = 100 km × 30 t × 0.00169 t.CO₂/(t·km) ≈ 5.08 t.CO₂

(3)

where ${\mathrm{D}}_{\mathrm{k}}$ is the transportation distance of the transport vehicle, with a value of 100 km; ${\mathrm{W}}_{\mathrm{k}}$ is the single trip/stage transportation volume, with a value of 30 t (the full load capacity of a single vehicle is 30 t); $\mathrm{E}{\mathrm{F}}_{\mathrm{k}}$ is the carbon emission factor of the transport vehicle. Referring to Guidelines for Calculating Carbon Dioxide Emission Intensity of Operating Vehicles in the Highway and Waterway Industry, the factor for heavy-duty diesel trucks (with a load of 30 t) is 0.00169 t.CO₂/(t·km).

4.

The carbon emission formula for component assembly is:

E₄ = P × t × EF_e = 15 kW × 16 h × 0.581 t.CO₂/(kW·h) = 1.39 t.CO₂

(4)

where P is the power of the assembly equipment. The main on-site equipment includes a 10 t small crane (10 kW) and small electric tools (5 kW), with a total power of P = 15 kW; t is the operation duration. According to the project records, the on-site assembly takes 2 days, with 8 working hours per day, and the actual equipment operation duration t = 16 h (excluding worker rest time); $\mathrm{E}{\mathrm{F}}_{\mathrm{e}}$ is the power carbon emission factor. Referring to the 2023 Regional Power Grid Baseline Emission Factors for CCER Projects, the national average power grid factor is 0.581 t.CO₂/(kW·h).

5.

The carbon emission reduction formula for clean energy is:

E₅ = −(G × EF_ee) = −(3568 kW·h × 0.581 t.CO₂/(kW·h)) = −32.11 t.CO₂

(5)

where G is the clean energy power generation (negative because it offsets the carbon emissions from grid power). The M-Box1.0 is fully covered with photovoltaic modules with a power of 2 kW, and the annual effective power generation hours are 1784 h, so G = 2 kW × 1784 h = 3568 kW·h; $\mathrm{E}{\mathrm{F}}_{\mathrm{e}\mathrm{e}}$ is the carbon emission factor of grid power (same as $\mathrm{E}{\mathrm{F}}_{\mathrm{e}\mathrm{e}}$), taking 0.581 t.CO₂/(kW·h)).

6.

The carbon emission formula for building system energy consumption is:

E₆ = ∑(E_m × EF_ee) = 5.32 kW·h/y × 0.581 t.CO₂/(kW·h) = 3.09 t.CO₂·y

(6)

where ${\mathrm{E}}_{\mathrm{m}}$ is the consumption of the m-th type of energy, such as electricity and gas. The M-Box1.0 has no gas equipment and only uses electricity. According to the operation records of M-Box1.0, the annual total electricity consumption ${\mathrm{E}}_{\mathrm{m}}$ = 5.32 kW·h/year (including electricity for air conditioning, lighting, and sockets); $\mathrm{E}{\mathrm{F}}_{\mathrm{m}\mathrm{e}}$ is the carbon emission factor of the corresponding energy. The power factor is the same as $\mathrm{E}{\mathrm{F}}_{\mathrm{e}}$ = 0.581 t.CO₂/ kW·h, and there are no other energy sources.

5. Conclusions

This study takes M-Box1.0 as a case to explore the design strategies of modular demountable building products for DFMA. By implementing four core strategies—modular design of the product system, rational design of assembly methods, optimized design of logistics and transportation, and integrated design of components—M-Box1.0 has achieved significant performance improvements. In terms of construction efficiency, only two workers can complete on-site assembly within 2 days. Economically, due to reduced material consumption and lower labor costs (with the manufacturing process transferred to the factory), the overall cost is only 61.1% of that of a steel structure building of the same scale. In terms of sustainability and environmental protection, the use of low-environmental-impact steel, modular recycling systems, and the “exoskeleton” structure not only further improves steel utilization efficiency through material reuse but also minimizes construction waste through prefabrication and demountable connections.

The innovations of this study are reflected in three aspects: first, it realizes the in-depth integration of DFMA theory and modular demountable buildings, transforming manufacturing-oriented design concepts into operable strategies such as disassembled sheet transportation and embedded angle bracket connections, breaking the barrier between architectural design and industrial production. Second, the proposed three-level modular classification system (module-component-part) and “exoskeleton” structure (integrating decoration, structure, and waterproof functions) solve the contradiction between manufacturing efficiency, assembly convenience, and structural stability in modular buildings. Third, it uses digital tools such as BIM, Inventor, and DFC to visually optimize the entire process of design-construction-logistics, avoiding conflicts in the manufacturing and assembly stages in advance and providing a new technical paradigm for building industrialization.

However, the study still has limitations: the current application of M-Box1.0 is mainly limited to small residential buildings and homestays, and its applicability in high-rise buildings or buildings with complex functions has not been verified. Future research should focus on:

• Expanding application scenarios to high-rise buildings, public buildings, and buildings in extreme environments, and testing the adaptability of DFMA strategies in complex structures;
• Establish a long-term performance monitoring system to track the structural performance, thermal insulation, and waterproof durability throughout the life cycle;
• Optimize recycling technology, develop intelligent disassembly equipment and component health assessment tools, and improve material reuse efficiency;
• Deepen digital integration, explore the combination of DFMA, parametric design, and machine learning, and realize the automatic optimization of modular systems.