Full length article
Product platform configuration for product families: Module clustering based on product architecture and manufacturing process

https://doi.org/10.1016/j.aei.2022.101622Get rights and content

Abstract

Product family design utilizes platform-based modularity to enable product variety and efficient mass-production. While product platform issues have attracted much attention from both academia and industry, traditional product platform design for product families emphasized the platform-based modularity that focuses on product structure dimension (functional or non-functional) to realize cost reductions during the design stage. Both the design architecture and manufacturing process are objectives that define product family modularity (PFM). They should be closely coupled with each other for the planning and configuration of platforms. This paper focuses on the product platform configuration by recognizing and utilizing shared product modules for product families. Instead of clustering product modules only based on their design structure, this approach differentiates each product variant, and considers the inherent relationship between product architecture and processing activities. The advantage is that similar components can be grouped and produced on a shared platform, thus benefitting from lower cost and shorter production time. First, both the architecture and manufacturing information of the product variety are captured in matrix format. Then, hierarchical clustering is applied over the components to generate PFM. Finally, a set of platforms are constructed to efficiently process most components of variants.

Introduction

Manufacturers increasingly move towards mass customisation to offer a higher level of product variety in response to the diverse demands of customers for different market segments [1], [2]. Product family design (PFD) was adopted by many companies to increase product variety, while achieving mass-production efficiency [3], [4], [5], [6], [7]. A product family refers to a set of similar product variants, derived from a common platform, that possess different features/functionality to meet specific and customized customer requirements [8], [9].

Product platform configuration of a product family is a key decision area each product development team faces during the early stages of product development [10]. A product platform consists of a series of modules and generates product variations by configuring existing modules [11], [12]. In platform-based PFD, a product is preferably designed and produced based on a platform by various design and manufacturing processes in a complicated context. Product variants are created on the product platform through configuration of existing modules, e.g., by addition, substitution, and/or removal of modules or components. Decision-making for configuration and planning of product platform for product families can be considered as a compound problem with respect to the incorporated communal modular design in the developmental stage with manufacturing process configuration at the production stage [13]. The main task of product platform designing is the identification of shared modules in correlative products. Product family modularity (PFM) architecture identification, based on product structure and manufacturing processes, are the most critical processes to support platform-based product family fulfilment.

The recent popularity of modular manufacturing models [14], [15] has enabled the utilisation of product variant structures that efficiently facilitate mass customised manufacturing. Platform-based modular design and manufacturing not only facilitates product development by translating customer requirements into product functions, but also enhances economics of scale [16], [17], [18]. Focusing on similar product variants for product families, PFM refers to the grouping or coupling of a set of similar common components into individual modules, mainly based on commonality criteria for the probability of component reuse and for the minimization of fulfilment cost in the product family. In particular, a set of shared components used on a product platform can be interpreted as a module or as a work-in-progress, depending on the required context of design, processing, and production sequence. Product families consist of vast versions of modules, subassemblies, components, and parts. This leads to an increase in variety through the design and production phase, and furthermore affects different processing and assembly levels of each product variant [19], [20]. Since the number of possible product variants can be high, identifying the PFM and configuring the product platform for product families is not a trivial problem.

Proliferation of product variation and the resulting variety levels across product structure in PFD affect all related manufacturing activities, particularly, processing and assembly [21], [22]. Vast research efforts have been devoted on how to build a product platform with a set of alternative modules. Most of the previous studies on the management of product varieties with configuration product platforms for modules mainly focused on similar efforts of product and component, and were based on the product structure (functional or non-functional) to meet diverse customer demands, while realising cost reductions during the design stage [23], [24], [25]. Existing methods of product platform configuration for product families considered product development and manufacturing process problems separately [26], [27], [28]. Prominent examples are that the PFM are specified a priori by designers on the front-end design of products, while the final product platform configuration decision is often made by manufacturers after receiving the PFM results at the back-end manufacturing stage. However, selecting the right combination PFM of common components and configuration of product platforms are not trivial because of the underlying conflicting goals and restrictions between design and manufacturing problems. In addition, PFM must consider the manufacturing activities and consequences of production in the manufacturing process. By considering the manufacturing process and/or assembly sequence for PFM, the release time of the modules placed on a product platform can be shortened, which consequently decreases related planning costs [29], [30]. Therefore, the platform-based PFM design should not only be applied during the design phase, but also should support manufacturing. The two problems (PFM design based on the design architecture and the manufacturing process) should be closely coupled with each other for the planning and configuration of product platforms for product families.

By focusing on PFM for platform-based problems, this paper explores the inherent joint mechanism for simultaneous consideration of product design and manufacturing process problems. In particular, PFM can be achieved from different perspectives of the product lifecycle, including issues such as development, manufacturing, and assembly. This means that the specific identification of PFM is needed to realize horizontal movement between design structures, manufacturing processes, and the incorporation of PFM into the configuration of platforms. There are three main stages of PFM design and configuration of platforms:

  • (1)

    Product architecture modularity. Architecture-based modularization typically describes a product structure using a product composition graph. From the perspective of product structure similarity, designers need to address the modules on the basis of structural coupling to identify subassemblies or modules.

  • (2)

    Manufacturing process modularity. During the manufacturing phase, manufacturing process formation, grouping, and platform-based modules are constructed as a standard of modularization toward realizing manufacturing efficiency [31], [32]. Manufacturing process modularity refers to the grouping of those components that require similar manufacturing processes into new modules, which can then be placed on the product platform.

  • (3)

    Configuration of platforms. For the configuration of platform stage, as shown in Fig. 1, by sharing modules and manufacturing processes across a platform, companies can efficiently develop differentiated products. This enables the clustering of product components and simplifies manufacturability, while also reducing planning complexity [33], [34]. Product family-driven modular design and platform configuration are by no means a simple ‘add-on’ approach that simply expands a single platform with additional PFM measures. At this stage, it is important to balance and standardise platforms during the manufacturing phases to realise cost savings [35], [36]. For example, if one of the platforms is overloaded or if material shortages occur, the cycle time increases, and hence, the throughput decreases [37], [38], [39].

To address the outlined problems, this paper simultaneously considers both objectives of product architecture and manufacturing process information to identify PFM and configure product platforms for product families. Mathematically, the platform-based PFM solution problem can be regarded as a grouping problem, and thus, cluster analysis can be considered for problem-solving [40], [41], [42], [43]. This paper uses the hierarchical clustering approach to address module and platform configuration problems. Specifically, as hierarchical clustering is applied as a non-exhaustive and ‘greedy’ solution strategy, the hierarchical relationships between different classes can be found without specifying the clustering number in advance. Besides, the proposed method is used to condense the membership relationship between modules and components into the relationship between parent and subclass. The method covers the complete process of platform configuration for product families, from the identification of component candidates at the design stage to their processing and assembly at the manufacturing stage. First, this study focuses on clustering product components to identify similar components and divides them into modules based on both the product variant architecture and manufacturing process for a series of variants. Furthermore, the manufacturing process can be assessed and rearranged for different modules where all platforms can be processed simultaneously.

The contributions of this paper can be described as follows:

  • (1)

    The coordinated configuration of platforms for product families simultaneously considers product architecture and manufacturing process information.

  • (2)

    Platform-based PFM clustering is designed to describe the information of design structures and the manufacturing process, and to incorporate PFM into the configuration of platforms. Grouped modules can be derived from the clustering method based on the variant-composition architecture. Furthermore, process activities can be rearranged based on the manufacturing process required by these grouped modules.

  • (3)

    Identification of a set of platforms for product families where shared components are used and processed to both shorten the production time and decrease the costs to facilitate production. The platforms are determined to ensure maximally smooth work station loads and material consumption of platforms.

The paper is structured as follows: Section 2 reviews the relevant literature. Section 3 provides an overview of the proposed method. Configuration and planning of product platforms based on the module clustering method and a numerical example to present the applicability for solving the platform planning problem are explained in Section 4. Section 5 provides a numerical example to verify the efficiency of the proposed method. Section 6 provides conclusions.

Section snippets

Optimal PFM solution for product families

To satisfy the increasing demands of customers for diversified products, many companies are developing PFD and product platforms to offer sufficient product variety. PFD offers a multitude of benefits, such as reduced development time, system complexity, and product costs, as well as the improved ability to upgrade products and disassemble them [43], [44], [45]. In the most general sense, modular design refers to the sharing of intellectual and material assets across a family of products.

Problem description

To illustrate the coordinated configuration process of platforms for product families under simultaneous consideration of the product architecture and manufacturing process information, the decision-making process between the design stage and the manufacturing stage are depicted in Fig. 1.

The designer designs the product structure based on customer requirements and specific decomposition criteria, such as functional or technological coupling. Product variant comprises several modules to realize

Application to configure the platform

This study uses data from a simplified design example of a valve to illustrate the proposed configuration platform process of family components of valves. This example is adopted from a case supported by the Metal forming Intelligent Machine Tool Co., Ltd., China. This enterprise designs and manufactures various types of hydraulic machines and produces moulding platforms based on specified requirements. A valve is mainly used as control part of the fluid transportation system for the machinery.

Cost analysis

To examine the economic implications of the results, the components data and cost data adapted from [41], [25] was used to illustrate the example. Here, the question is how to configure the platform to reduce production costs. This study observes whether the costs of designed platforms will change if component modules are constructed by both the product architecture and the process flow.

The cost of components is formulated as:TAc=(tcp+tca+tcr)dci

The cost of processing components (TBc) in a

Conclusions

Platform-based configuration methods for a specific product family are increasingly applied in many industries to meet the growing trend for increased product variability, diversification of customer demands, shorter life cycles, and cost reduction to facilitate mass customisation. Based on the basic information of product variety, the component architecture and processing information are specified in a matrix format, which captures the relationships between product design and the manufacturing

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was supported by grants from the National Natural Science Foundation of China Nos. 72171069, 71901086 and 72101075), and the Fundamental Research Funds for the Central Universities (JZ2021HGTA0128 and JZ2021HGQA0209).

References (61)

  • S. Venkatesh et al.

    Managing Product Variety Through Postponement: Concept and Applications

    (2004)
  • N. Tuptuk, S. Hailes, Security of Smart Manufacturing Systems,...
  • T. Agrawal et al.

    A hybrid model of component sharing and platform modularity for optimal product family design

    Int. J. Prod. Res.

    (2013)
  • Z. Pirmoradi, G.G. Wang, Recent advancements in product family design and platform-based product development: a...
  • T. Blecker et al.

    Mass customization: state-of-the-art and challenges

    Mass Custom. Chall. Solut.

    (2006)
  • H. Zhang et al.

    Progressive modelling of feature-centred product family development

    Int. J. Prod. Res.

    (2019)
  • R. Wang et al.

    A process knowledge representation approach for decision support in design of complex engineered systems

    Adv. Eng. Inf.

    (2021)
  • T.W. Simpson, J. Jiao, Z. Siddique, K. Hölttä-Otto, Advances in Product Family and Product Platform Design,...
  • S. Huang, Y. Yan, Design of delayed reconfigurable manufacturing system based on part family grouping and machine...
  • J. Fang et al.

    A knowledge support approach for the preliminary design of platform-based products in Engineering-To-Order manufacturing

    Adv. Eng. Inf.

    (2020)
  • T.W. Simpson et al.

    Product platform design: method and application

    Res. Eng. Des.

    (2001)
  • O.L.D. Weck, Product Platform and Product Family Design: Methods and Applications,...
  • C. He et al.

    A systematic data-mining-based methodology for product family design and product configuration

    Adv. Eng. Inf.

    (2021)
  • K. Baylis et al.

    Product family platform selection using a Pareto front of maximum commonality and strategic modularity

    Res. Eng. Des.

    (2018)
  • J. Stjepandic, A.J. Fougères, M. Kurth, E. Ostrosi, Modularity and Supporting Tools and Methods, Concurrent Engineering...
  • T.D. Brunoe et al.

    Integrated product-process modelling for platform-based co-development

    Int. J. Prod. Res.

    (2019)
  • F.G. Galizia et al.

    Product platforms design, selection and customisation in high-variety manufacturing

    Int. J. Prod. Res.

    (2019)
  • M. Aydin et al.

    A new methodology to cluster derivative product modules: an application

    Int. J. Prod. Res.

    (2016)
  • J. Cha et al.

    Introduction to the special issue 'Concurrent Engineering'

    J. Aerosp. Oper.

    (2015)
  • H.L. Lee et al.

    Designing Products and Processes for Postponement

    (1994)
  • Cited by (12)

    View all citing articles on Scopus
    View full text