Research on Building Customized Furniture Product Model and Assembly Sequence

Yue Qi, Yang SunInternational Journal of Smart Home 2015 · DOI

Abstract

With the improvement of people's living standard, mass customization production model has been the trend of the furniture industry. A model of furniture product design and assembly for customization is proposed. It can decompose customers' needs and transform them into homologous modules set in the structure tree. Then through the assembly relationship graph and matrix of these product parts, an assembly sequence which can guide the assembly of products will be built by a genetic algorithm. This process will not only save manpower and resources, but also improve enterprise efficiency. In this article, an instance was given in order to verify the correctness of product design and assembly processes based on product family of mass customization.

Introduction

In the traditional furniture industry, there is a significant lag between shifting customer demands and production capabilities. As "mass customization" becomes the gold standard—providing personalized products at near mass-production efficiency—manufacturers face a bottleneck: how to translate a customer's unique wishlist into a concrete manufacturing plan without manual engineering for every single order. This paper addresses the challenge of "variant design," bridging the gap between the Functional Domain (what the customer wants) and the Structural Domain (how the furniture is built), while ensuring that the resulting custom product can be assembled in the most efficient way possible.

Methodology

Function-Structure Mapping

The researchers utilize axiomatic design theory to decompose a product into four domains: customer, functional, physical, and process. By creating a Functional Tree (e.g., "convenience" or "placement") and mapping it to a Structural Tree (e.g., "handles" or "clapboards"), the system formalizes the transition from abstract desires to physical components. This hierarchical decomposition allows for modular variety while maintaining structural integrity.

Decomposition of TV cabinet function and structure trees.

Generalized Module Matrix

To handle variety, product components are organized into a matrix. Rows represent "module chains" (parts that work together), while columns represent variations in parameters such as size, material, or finish. This structure allows the system to select specific variants based on the functional requirements identified in the previous step.

Assembly Relationship Modeling

The physical connections between parts are modeled as an undirected graph. If two parts touch or connect, they share an edge. This graph provides a visual and mathematical representation of the assembly constraints, which is then converted into a matrix for computational analysis.

Assembly relationship graph for the TV cabinet components.

Genetic Algorithm (GA) Optimization

The core of the assembly planning is a Genetic Algorithm. The algorithm treats a potential assembly sequence as a "chromosome." It evaluates these sequences based on an adaptive fitness function that considers:

  • Assembly Direction ($f_1$): Minimizing the number of times a worker has to flip or rotate the piece.
  • Stability ($f_2$): Ensuring the partially assembled piece doesn't fall apart during the process.
  • Geometric Feasibility ($f_3$): Checking for physical interference (e.g., ensuring internal shelves are placed before the final enclosure).

Case Study: Customized TV Cabinet

The paper validates the model using a specific case study of a customized TV cabinet. The instance involves a modern-style cabinet with bamboo texture, visible glass doors, and specific internal shelving requirements. The primary structural set consists of 8 parts: apical board, bottom board, back board, two side plates, two vertical boards, and a clapboard.

The Genetic Algorithm evolved through several generations to produce an optimal assembly sequence for these 8 core structural parts: 7 → 1 → 3 → 4 → 6 → 8 → 5 → 2.

Discussion and Key Insights

The proposed framework formalizes the transition from vague customer desires to physical parts. By breaking "Appearance" down into "Style" and "Texture," and then mapping those to "Pattern" and "Fabric" modules, the authors create a repeatable logic for customization.

Furthermore, by treating assembly as a sequence optimization problem, the research demonstrates that even for a simple 8-part cabinet, there are thousands of possible sequences, but only a few are truly efficient. A key insight in the fitness function is the mathematical representation of stability, which ensures that each added part has maximum "connectivity" to the existing structure, preventing precarious builds during the manufacturing process.

Limitations and Future Work

While the framework is robust, the authors acknowledge several areas for growth. The paper focuses on the logic of design and sequence but does not deeply explore the specific manufacturing technologies (like CNC settings) required to execute the plan. Additionally, there is a lack of an intuitive, visual interface that would allow a non-technical customer to interact with the Functional Tree in real-time. Finally, future research is needed to see how the Genetic Algorithm performs as the part count increases exponentially for complex, multi-room furniture systems.

Summary updated

Abstract

With the improvement of people's living standard, mass customization production model has been the trend of the furniture industry. A model of furniture product design and assembly for customization is proposed. It can decompose customers' needs and transform them into homologous modules set in the structure tree. Then through the assembly relationship graph and matrix of these product parts, an assembly sequence which can guide the assembly of products will be built by a genetic algorithm. This process will not only save manpower and resources, but also improve enterprise efficiency. In this article, an instance was given in order to verify the correctness of product design and assembly processes based on product family of mass customization.
BibTeX
@article{qi2015research,
  author = {Yue Qi and Yang Sun},
  title = {{Research on Building Customized Furniture Product Model and Assembly Sequence}},
  journal = {International Journal of Smart Home},
  year = {2015},
  doi = {10.14257/ijsh.2015.9.3.08},
}
Figure