Abstract

The three-dimensional (3D) design data employed in a process plant construction project are generated during both the basic design and detailed design stages and are used for various purposes throughout the life cycle of the project. After the design stage, 3D design data are converted to a lightweight 3D format and utilized to support procurement, construction, and audit work in a collaborative project management system. However, significant time and cost are incurred when separate interfaces to convert design data are developed for each plant 3D computer-aided design (CAD) system. As an alternative, a method exists to integrate an interface using a neutral model. After translating the 3D input design data for the plant 3D CAD system to a neutral format, this study proposes an interface for use in collaborative project management by converting the data into a lightweight 3D model. In addition, detailed techniques for implementing the proposed interface are described. To verify the validity of the proposed neutral model-based 3D design data interface, translation, inspection, and lightweighting experiments are performed using 3D design data for a synthesized natural gas production plant project.

Highlights
  • A new interface is proposed to convert 3D design data to a lightweight 3D model.

  • The neutral 3D design model was defined to represent 3D computer-aided design and catalog data.

  • Plant 3D design data were converted into neutral model data.

  • Consistency and accuracy of the neutral data were checked by integrated visualization.

  • The neutral data were converted into a lightweight model tailored for large-scale facilities.

1. Introduction

With recent advances in geometric modeling, computer graphics, and computing power, the manufacturing industry is now utilizing a wide range of three-dimensional (3D) models in all engineering fields, including design, manufacturing, and maintenance. Even in a process plant construction project, 3D design data are used as the master data throughout the project period. The 3D design data also constitute the primary information that engineering, construction, and procurement companies must hand over to the operator to support operation and maintenance after plant construction.

The 3D design data created in the basic design and detailed design stages are stored in the native format of the plant 3D CAD system. After the design stages, a collaborative project management system is utilized to support collaboration with external organizations and manage internal business processes during procurement, construction, and audits. The collaborative project management system provides various functions, including project data integration, reviewing, and collaboration as well as engineering property checking, dimensional measurement, interference, and collision checking, and 2D drawing generation. This collaboration function is implemented based on the 3D design data of the corresponding plant.

The 3D design data used in the collaborative project management system have a general-purpose lightweight 3D format, rather than the native format of the plant 3D CAD system. Large complex facilities, such as process plants, generally consist of millions or more parts. If a 3D CAD model that requires more data than other shape representation methods is used, then more space is required to store the 3D design data, and data processing consumes more computing power. Typically, a general-purpose lightweight 3D model viewer incurs lower operating costs than a 3D CAD system. Furthermore, if the 3D CAD model is used in collaboration process, the company’s intellectual property, which is explicitly or implicitly included in the 3D CAD model, could leak to external partners. Therefore, an interface that converts the 3D design data generated by the plant 3D CAD system into a general-purpose lightweight 3D model is required to support plant project collaboration management.

Several types of systems are used for process plant 3D design. One method involves building an interface for each system for direct conversion of the 3D design data into a lightweight 3D model. However, this requires significant time and cost during the development stage, which is disadvantageous in terms of maintenance. An alternative method is to extract the design data required for collaborative work from the plant 3D CAD system in a neutral format and use these data for the collaborative project management system. In this case, it is sufficient for the collaborative project management system to provide an interface for the neutral model without considering the interface for all plant 3D CAD systems. Thus, this method is advantageous in terms of separating the plant 3D design and project collaboration stages, as well as for the development and maintenance of the data conversion interface. Besides, external data interfaces of plant 3D CAD systems are boundary representations (B-rep) or mesh formats. B-rep models have relatively large data size and thus are considered to be lightweight models. Mesh models exported directly from a plant 3D CAD system also have large data size if they do not go through additional lightweighting processes. Therefore, after plant 3D design data are translated into neutral model data, additional lightweighting processes should be performed in 3D CAD models comprising the translated neutral model data.

This study proposes a method to extract the 3D CAD model (i.e. the 3D design data of the process plant, along with the equipment and material catalog) in a neutral format and convert it to a lightweight 3D model for use in project collaboration. For the translation of design data generated from the plant 3D CAD system, a neutral model is defined by referring to the relevant industrial data standards. The consistency and accuracy between the 3D CAD model and the catalog constituting the 3D design data are further checked. The translated neutral 3D design data are then converted into a lightweight 3D model and used in a collaborative project management system. Finally, a prototype system is developed, and the proposed method is verified by translating the test 3D design data, converting it into lightweight format data, and conducting collaborative work experiments.

This study provides three academic contributions compared to previous studies. First, a neutral 3D design model is proposed to ensure the link between the 3D CAD model and catalog data that comprise plant 3D design data. Second, since large complex facilities, including process plants, comprise an enormous number of parts, additional lightweighting techniques are applied as well as the conventional conversion to triangular mesh. Third, considering the proposed neutral 3D design model and additional lightweighting, this study proposes a method consisting of two steps: translating plant 3D design data into neutral model data and converting neutral model data to a lightweight 3D model.

The remainder of this paper is structured as follows. Section 2 reviews previous studies on lightweighting 3D CAD models and plant 3D design data exchange. Section 3 proposes a neutral model-based 3D design interface. Section 4 presents the translation of plant 3D design data into neutral model data. The visualization-based verification of neutral 3D design data is presented in Section 5, and the conversion to a lightweight 3D model is presented in Section 6. Section 7 discusses the results of implementation and experiments. Conclusions and suggestions for future work are provided in Section 8.

2. Related Work

Methods to reduce the data size of 3D CAD models can be classified as model simplification (Kim & Mun, 2014a) or model lightweighting (Kwon & Mun, 2019). Model simplification reduces the data size of the 3D CAD model using an evaluation metric to calculate the importance of the various shape elements involved. Then, a simplification operation is applied to sequentially remove shape elements with low importance and to fill in any resulting voids until the required level of detail (LOD) is achieved (Kwon et al., 2015). LOD is a term commonly used in computer graphics and represents the degree of detail of the shape of a 3D model. The specific method used to simply a 3D CAD model depends on the representation methods of the model (Kwon et al., 2019). For example, 3D CAD models that take the B-rep form (Pratt et al., 2005) can be simplified by analysing the composition pattern of the topological elements constituting the model and then identifying and removing target elements to be simplified (e.g. faces; Koo & Lee, 2002; Kim et al., 2005; Sun et al., 2010) or by performing volume decomposition, creating a volume list, and then applying feature-based simplification (Woo, 2009; Kim & Mun, 2014b, 2015).

By contrast, model lightweighting reduces the data size of a 3D CAD model format, such as B-rep, by converting it to a 3D lightweight model format, such as Jupiter tessellation (JT; JT File Format, 2020), stereolithography (STL; STL file format, 2020), or wavefront object (OBJ; OBJ file format, 2020). Typically, the lightweight model represents the shape via a triangle mesh, and the size of the lightweight file depends on the triangle structure to be stored, the storage method (text or binary), and the compression method. In addition to representing 3D shapes with triangles, the file size can be reduced by using a hybrid form of constructive solid geometry primitives, e.g. the hybrid method proposed by Nguyen and Choi (2019). Eigner et al. proposed a method to increase the utilization of a lightweight model by combining a JT file and an extensible markup language (XML) file containing additional information (Eigner et al., 2010). More recently, Kwon and Mun proposed a method to reduce the size of a lightweight model by categorizing the types of parts that make up the structures of ships and offshore plants, and then storing the minimal triangle mesh for each part type without storing unnecessary information (Kwon & Mun, 2020).

Large complex facilities, including process plants, consist of many parts, which lead to the necessity of adopting additional lightweighting techniques. First, when converting the 3D shape of a part having the B-rep format into a triangle mesh, if the shape elements’ pattern is analysed and corresponds to a primitive, it is converted and stored as a parameter value list rather than a triangle mesh. In addition, the concept of a mesh block was introduced for swift rendering. Thus, by grouping the 3D shapes of end nodes sharing the same visual properties and processing them using a mesh block, the number of shape transmissions is significantly reduced. A lightweight model data structure was newly defined to cover new techniques applied in this study.

The use of standards to manage facility asset information in the process plant industry has been investigated by Braaksma et al. (2011). They reported that, in practice, information standards are applied only to a limited extent. The primary international standards used for the exchange of design information in process plants include ISO 10303 (Owen, 1997), ISO 15926 (Leal, 2005), and ISO 13584 (Cho et al., 2006). ISO 10303 deals with the exchange of product model data and provides an application protocol (AP) applicable to the process plant industry. ISO 15926 addresses the sharing and integration of process plant information, and ISO 13584 is relevant to the representation and exchange of parts library data. The published research on the exchange of standard-based equipment specification information indicates that an equipment and material catalog system based on ISO 15926 was established and used in ship and offshore plant construction projects in the European Union (Irgens et al., 2004; qHub, 2016). For example, Kim et al. implemented an ISO 15926-based data storage prototype called a facade to store the equipment and materials data of nuclear power plants and to provide the data for related organizations (Kim et al., 2011). More recently, Kwon et al. proposed a method to improve the sharing environment of catalogs for equipment and materials by representing specifications data using ISO 15926 (Kwon et al., 2016). In addition, Fiorentini et al. conducted a study to convert existing engineering data of nuclear power plants to the ISO 15926 format (Fiorentini et al., 2013).

The ISO 15926 standard has also been used in studies on the exchange of standards-based 3D CAD models and 2D drawings. For example, Kim et al. proposed a method to exchange plant 3D design models (Kim et al., 2017), while Li et al. referred to both ISO 15926 and ISO 10303 AP 227 to develop a neutral model for the exchange of 3D design models for ship outfitting between the Tribon design and information system and the plant design management system (PDMS; Li et al., 2011). Kim et al. proposed a method for exchanging procedurally represented 2D drawing data using part 112 of the ISO 10303 standard (Kim et al., 2011).

With respect to the integration of standard-based process plant life cycle information, Lee et al. applied the basic concepts of ISO 15926 part 2 to propose a data model that supports the effective operation and maintenance of large complex facilities (Lee et al., 2012). More recently, Kim et al. proposed a data model based on ISO 15926 that can link the life cycle data of a process plant including maintenance activities (Kim et al., 2020). The data model proposed by Kim et al. provides information resources that can link facility change history to design, manufacturing, and installation information according to maintenance needs. Kim et al. also proposed a system architecture to generate, store, and manage the neutral catalog of components together with a capability to translate the neutral catalog into the native format of a plant 3D CAD system (Kim et al., 2021).

Previous studies of plant 3D design data exchange generally fall into two categories: the exchange of 3D CAD models (Li et al., 2011; Kim et al., 2017; Safdar et al., 2020) and the exchange of catalog data (Lee et al., 2012; Kim et al., 2020). To the best of knowledge, no studies have investigated a method to translate plant 3D design data that ensures the link between the 3D CAD model and catalog data that comprise plant 3D design data. This study proposes a method to translate both 3D CAD model data and catalog data, ensuring the link between them. In addition, the proposed method defines a neutral model used for this translation.

3. Configuration of Neutral Model-Based Interfacing of 3D Design to Project Management

Detailed design in the plant 3D CAD system proceeds by searching the equipment and material catalog library, selecting the required plant items, and placing them in a 3D space for each individual discipline, such as piping, equipment, structure, and electricity, as well as heating, ventilation, and air conditioning (HVAC). The 3D CAD model contains information on the assembly relationship, connection relationship, general properties, and 3D arrangement (position and rotation) of equipment and materials. Furthermore, the catalog used for modeling in this manner is stored and managed by the plant 3D CAD system along with the resulting 3D CAD model in a separate database. The catalog contains information on the 3D shape, specifications, and ports of the equipment and materials.

As shown schematically in Fig. 1, considering the data items managed by the plant 3D CAD system, a neutral model-based 3D design data interface is then defined to support the project management of a process plant. The 3D design data interface consists of a 3D CAD model translation unit, a catalog translation unit, a verification unit, and a conversion unit, i.e. conversion to lightweight 3D models. The 3D CAD model translation unit converts the model for each discipline created in the plant 3D CAD system into a neutral format. The catalog translation unit converts the equipment and material catalog into a neutral format, which is referenced in the plant 3D CAD system for 3D design. The neutral model representing the 3D design data consists of two submodels representing a 3D CAD model and a catalog based on the ISO 10303 and ISO 15926 standards. The 3D CAD model is defined by reference to the information resources provided by ISO 10303 AP 227, and the catalog model is defined by reference to the equipment and material specifications in ISO 15926. The two submodels are primarily interlinked via the catalog ID. The verification unit then checks the consistency and accuracy of the translated neutral 3D design data via integrated visualization. In terms of data consistency, this unit checks whether the catalog referenced by equipment and materials constituting the 3D CAD model is provided in the catalog data. In terms of data accuracy, this unit checks whether the information stored in the neutral 3D design data matches the information generated by the plant 3D CAD system. Then, the conversion unit converts the neutral 3D design data into a lightweight 3D model. The neutral 3D CAD model is converted directly into a lightweight 3D model, while the neutral catalog is converted into an annotation data format for use as an auxiliary support for the lightweight 3D model. As with the neutral 3D design data, the lightweight 3D model and annotation data are primarily linked via the catalog ID. The converted lightweight 3D model is then used in the collaborative project management system for project data integration, project reviewing and collaboration, dimensional checking, interference checking, and crosssection visualization.

Configuration of the neutral model-based 3D design interface.
Figure 1:

Configuration of the neutral model-based 3D design interface.

4. 3D Design Data Translation in a Neutral Model

As described above, a neutral 3D design data model was defined to translate the native 3D design data generated from the plant 3D CAD system into a neutral format. The neutral model of the 3D design data consisted of a neutral 3D CAD model and a neutral catalog model. Native 3D design data were translated into neutral 3D design data according to the method proposed in this study.

Different plant 3D CAD systems have different data structures. Therefore, in this study, a neutral 3D CAD model and a neutral catalog model were defined by applying the concept of reference data in ISO 15926 (Leal, 2005). The basic principle is defining only classes, properties, and objects that are common in the target domain in the explicit structure of the model and representing detailed model information using separate reference data.

For example, in the catalog model described in Section 4.2, a class object is defined within the model. In addition, reference data for component and equipment classification, such as a valve, elbow, cap, and tee, are declared instances of class objects in auxiliary files. Accordingly, when translating catalog data, data translation is performed using the mapping relationship between the classification of components and equipment of the commercial system and the classification of components and equipment defined in the reference data.

4.1. 3D CAD model translation

The neutral 3D CAD model is defined as shown in Fig. 2, where the Plant object refers to the entire plant and a System is provided for each discipline. Among several possible systems, this study deals exclusively with equipment and piping. Thus, the Plant object is subdivided into a Piping System object and a Mechanical System object (Fig. 2). The Piping System object stores the piping design results. Here, the Piping object refers to a set of pipes that perform a specific function and represents the connection information between one Piping Segment and another via the Relation object. Each Piping Segment represents a branch, where branches identify points at which a flow diverges or changes. The hierarchical information of Piping Segment and Piping Component is also represented via the Relation object. The Piping Component object represents a plant item for fittings and has the component specification (CSPEC) property with respect to catalog reference (CATREF) information. The Mechanical System object is a collection of equipment that contains design information regarding individual equipment. The connection information between Equipment and Piping Component is also represented via the Relation object. In addition, Nozzle objects represent nozzles that connect equipment and piping. The Nozzle object has a CATREF property with respect to CATREF information.

Data structure of neutral 3D CAD model.
Figure 2:

Data structure of neutral 3D CAD model.

Objects under Plant have common Object Properties. The Object Property contains object property information, i.e. Port, Id, Position, Spec, Insulation Spec, and Sat. Here, Port is a characteristic of the Nozzle, Piping Segment, and Piping Component that indicates connection point information between pipes and equipment or between individual pipes. The Id property represents unique identification information and is a characteristic of all objects (except Plant). The Position property indicates position information and is a characteristic of Equipment and Piping Component. The Spec property represents a set of catalog selection rules that satisfy given design requirements and is a characteristic of Piping and Pining Segment. The Insulation Spec property indicates information related to heat transfer (insulation) treatment and is a characteristic of Nozzle. The Sat property is a characteristic of Equipment and Piping Segment that represents the address of the standard ACIS text (SAT) file in which the plant 3D shape information is stored.

As shown in Fig. 3, the native 3D CAD model is input to the 3D CAD model translation module and translated to a neutral 3D CAD model via (i) shape information extraction, (ii) schema mapping, and (iii) 3D CAD model data translation. Thus, the 3D shape information in the native 3D CAD model is first extracted and stored as an SAT format file. Commonly used neutral 3D CAD model formats include IGES, STEP, SAT, Parasolid-XT, JT, and STL. Plant 3D CAD systems, e.g. AVEVA PDMS and HEXAGON Smart3D, support the SAT format. Thus, the SAT format was adopted to store 3D shapes in the neutral model. As a result, the correspondence between the constituent objects of the 3D CAD model of the plant 3D CAD system and the neutral 3D CAD model is defined via schema mapping. For the Piping Component object, the mapping relationship varies depending on the type of fitting; thus, a separate mapping file was defined manually and utilized in the mapping process. In the 3D CAD model data translation process, the native 3D CAD model data are primarily read, and their components are converted to those of the neutral 3D CAD model using the mapping information stored in the mapping file. In the 3D CAD model, there are three mapping files, i.e. the class file (stores classes comprising an assembly relationship), the member file (stores components and equipment belonging to each class), and the attribute file (stores a list of attributes for each member). The results are stored in the internal data structure and are output as neutral 3D CAD model data in the XML format.

Procedure to translate 3D CAD model data.
Figure 3:

Procedure to translate 3D CAD model data.

4.2. Catalog translation

The neutral catalog model is defined as shown in Fig. 4. Here, the object Spec represents a set of catalog selection rules that satisfy the given design requirements, the Selection Filter Class object defines a list of attributes required to select a catalog according to the specifications (e.g. nominal diameter) and type of equipment and material (e.g. fitting, gasket, or bolt), and the Selection Filter represents the selection rules for individual catalogs according to the list of attributes. In addition, the Attribute object represents the attributes of the Selection Filter Class object, and its values are represented as Attribute Value objects. The Class object represents types of equipment and materials, and the Property object represents the properties of a specific equipment or material type, and its values are represented as the Property Value object. The Catalog object stores property values representing the functional and physical specifications of specific equipment or material. The Catalog object stores the property value of the corresponding class object in the Property Value object format. The types of equipment and materials represented by the catalog are identified depending on a Class object. The Related Class relationship of the Class object and Child Catalog relationship of the Catalog represent information about the hole or nozzle in the equipment. In other words, the equipment catalog and nozzle catalog are linked through the Child Catalog relation, and the Related Class relation represents the relationship between nozzle type and hole type.

Data structure of neutral catalog model.
Figure 4:

Data structure of neutral catalog model.

As shown in Fig. 5, the native catalog model is input to the Catalog translation module and translated into a neutral catalog via schema mapping and catalog data translation. Here, schema mapping defines the correspondence between Spec, Class, Selection Filter, Property, Attribute, and Code for the catalog model of the plant 3D CAD system and the neutral catalog model. The neutral catalog model is stored and used in an auxiliary file. In addition, the manually defined mapping relation is stored in a separate mapping file and utilized in the translation process. In the catalog translation process, the neutral catalog model is loaded first, the components of the native catalog are converted to those of the neutral catalog using the mapping information stored in the mapping file, and the result is stored in the internal data structure. Finally, the result is output as neutral catalog data in the XML format.

Procedure to translate catalog data.
Figure 5:

Procedure to translate catalog data.

5. Visualization-Based Verification of Data Consistency and Accuracy in Neutral Model

The neutral 3D CAD model and neutral catalog are converted independently from the plant 3D CAD system; thus, it is necessary to verify the consistency between these two datasets. Thus, prior to the lightweighting process, the neutral 3D design data are subjected to a visualization-based verification process, where the data consistency and accuracy are evaluated by integrated visualization of the translated neutral 3D design data (Fig. 6). Specifically, the process determines whether the catalog information for each plant item constituting the neutral 3D CAD model is provided from the neutral catalog. In addition, this process verifies whether the 3D CAD model and catalog information are translated accurately.

Integrated visualization procedure for neutral 3D design data.
Figure 6:

Integrated visualization procedure for neutral 3D design data.

For this verification, the neutral 3D design data are loaded and then visualized in 3D. Consistency between the 3D CAD model and catalog is then further verified as follows. In the case of a piping design, the catalog ID stored in the CSPEC property of the Piping Component object is extracted, and the presence of a catalog item with the corresponding ID is checked for in the neutral catalog. In addition, the equipment or material type properties recorded in the catalog are extracted to determine whether they match the type of Piping Component object. For equipment design, the catalog ID stored in the CATREF property of the equipment is extracted, and the presence of a catalog with the corresponding ID in the neutral catalog is checked. After verifying data consistency, data accuracy is verified by checking the following aspects: (i) whether the major information of the neutral 3D CAD model (e.g. assembly relationship, 3D shape, and nonshape property) is displayed without issue; (ii) whether the displayed data exhibit discrepancies compared to the native 3D CAD; (iii) whether the major information in the neutral catalog is displayed without issue; and (iv) whether the displayed data exhibit discrepancies compared to the native catalog.

6. Conversion to Lightweight 3D Model with Annotations

To convert the neutral 3D design data to a lightweight model, the data structure of the lightweight 3D model is defined as outlined in Fig. 7. This includes product manufacturing information, structure, shape, material, and user-defined attributes. The product structure contains information about the hierarchical relationship between parts (equipment and materials) constituting the product and is represented using the Assembly, Part, and Instance objects. The Body object is connected to the Part object and contains detailed shape information for each part, including different phase information (i.e. Face, Loop, Edge, and Vertex). Here, the detailed shape is represented as a Mesh Block object (i.e. a set of triangle sets) or a Triangle Set object in the form of a triangle mesh. Note that a wireframe model has an advantage over a mesh model in terms of data size; however, using a wireframe model, it is difficult to satisfy all requirements of a collaborative project management system based on a lightweight 3D model. For example, if a wireframe model is used, there is a limit to implementing functions, e.g. construction quantity calculation and interference check. Thus, triangle meshes are primarily used. The geometries corresponding to the Face, Edge, and Vertex are represented by a nonuniform rational B-spline (NURBS) Surface, NURBS Curve, and Point objects, respectively.

Data structure for lightweight 3D models.
Figure 7:

Data structure for lightweight 3D models.

Each part corresponding to an end node in the product structure includes 3D shape information, and the 3D shape video must be transmitted to video memory during rendering. Large complex facilities, e.g. process plants, comprise millions of parts; thus, it is difficult to ensure sufficient visualization performance when rendering is performed while traversing the product structure. To solve this problem, the mesh block concept is introduced to realize fast rendering. Thus, by grouping the 3D shapes of end nodes sharing the same visual properties and processing them using a mesh block, the number of shape transmissions is reduced significantly. Typically, visually distinguishable 3D CAD models are created by assigning one of 16 or 256 colors to each part; thus, even for an assembly containing thousands of parts, using a mesh block enables suitable rendering via 16 or (at most) 256 shape transmissions.

The lightweight 3D model is converted from a neutral 3D CAD model that is part of the neutral 3D design data. The product structure of a lightweight 3D model is then constructed from the assembly and connection relationships stored in the neutral 3D CAD model. After constructing the product structure, the 3D shape of the part is transformed. In the neutral 3D CAD model, the 3D shape is stored as a B-rep format SAT file. During the conversion process, the neutral 3D CAD model is converted to a triangle mesh and stored in the Triangle Set object of the lightweight 3D model. The general properties of the neutral 3D model are then converted to user-defined attributes of the lightweight 3D model. Although neutral catalog data are not converted separately, the catalog ID referenced by each part in the Attribute object of the lightweight 3D model is stored to allow access of the lightweight 3D model to the neutral catalog.

In a plant 3D CAD system where modeling is performed using a catalog, individual parts, e.g. pipes, fittings, and structural members, can occasionally have the same shape. In such cases, after creating one Part object for the same part, the Instance object that refers to this Part object can be used to create a product structure. Note that the Instance object only has location and direction information; however, the 3D shape information is obtained by accessing the referenced Part object. In plant 3D CAD systems, a combination of primitives (e.g. cylinders and spheres) is often used to represent 3D shapes. Here, file size can be reduced effectively by storing these primitives as a list of required descriptive parameter values rather than storing them in a triangle mesh. When converting the 3D shape of a part with the B-rep format into a triangle mesh, if shape elements are analyzed as corresponding to a primitive type, they are converted to a primitive rather than a triangle mesh, storing a parameter value list of the primitive.

7. Implementation and Experiments

To verify the proposed neutral model-based 3D design data interface, a prototype system was developed according to the above method. This prototype comprises four modules with the implementation environments listed in Table 1. Here, the 3D CAD model translation and catalog translation modules utilized the available programmable markup language (PML) to extract the unique 3D design data from the PDMS, and the neutral 3D design data verification module used the commercial InterOP, ACIS, and HOOPS3D libraries for conversion, processing, and visualization of the 3D shape files in the SAT format. In addition, the conversion unit to lightweight 3D models used InterOP to read the 3D shape files in the SAT format.

Table 1:

Implementation environments of modules comprising prototype system.

UnitsModulesEnvironments
3D CAD model translation unit3D CAD model translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Catalog translation unitCatalog translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Verification unit of neutral model-based 3D designNeutral 3D design data verification module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: ACIS, InterOp, Hoops3D
Conversion unit to lightweight 3D modelsLightweight 3D model conversion module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: InterOp
UnitsModulesEnvironments
3D CAD model translation unit3D CAD model translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Catalog translation unitCatalog translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Verification unit of neutral model-based 3D designNeutral 3D design data verification module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: ACIS, InterOp, Hoops3D
Conversion unit to lightweight 3D modelsLightweight 3D model conversion module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: InterOp
Table 1:

Implementation environments of modules comprising prototype system.

UnitsModulesEnvironments
3D CAD model translation unit3D CAD model translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Catalog translation unitCatalog translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Verification unit of neutral model-based 3D designNeutral 3D design data verification module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: ACIS, InterOp, Hoops3D
Conversion unit to lightweight 3D modelsLightweight 3D model conversion module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: InterOp
UnitsModulesEnvironments
3D CAD model translation unit3D CAD model translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Catalog translation unitCatalog translation module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C#
- Libraries: PML
Verification unit of neutral model-based 3D designNeutral 3D design data verification module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: ACIS, InterOp, Hoops3D
Conversion unit to lightweight 3D modelsLightweight 3D model conversion module- OS: MS Windows 10 pro 64 bit
- CPU: Intel Core i7 | RAM: 64.00 GB
- Language: C++
- Libraries: InterOp

Then, an experiment was performed in which the prototype system was used to convert the 3D design data of the synthesized natural gas production plant modeled in the field to a lightweight 3D model, and the data were applied in the collaborative project management system. Here, the experimental data were the 3D design data of the Synthetic Natural Gas (SNG) plant completed in 2016 in Gwangyang, Jeollanam-do, South Korea. As shown in Fig. 8, the data used in this experiment were modeled using the AVEVA PDMS, the piping 3D design data comprised 5226 components, and the equipment 3D design data comprised 616 pieces of equipment. Note that these data also included a catalog of 616 components and equipment.

Experimental test data.
Figure 8:

Experimental test data.

The process and results of converting the 3D design data into a lightweight 3D model are shown in Fig. 9. First, the native 3D CAD model and catalog were extracted from the PDMS using the 3D CAD model translation module and the catalog translation module, and these were then translated into neutral model data. The translation proceeded according to the method described in Section 4, and the results are shown in Fig. 9a. This neutral model translation method has an advantage; i.e. it can be applied equally to the native 3D design data of other plant 3D CAD systems with different data structures. For example, the result of translating the 3D design sample provided by HEXAGON Smart3D into neutral model data is shown in Fig. 9d.

Neutral model-based interfacing of 3D design data to lightweight 3D model.
Figure 9:

Neutral model-based interfacing of 3D design data to lightweight 3D model.

A visualization-based verification was performed on the 3D design data in the neutral format using the neutral 3D design data verification module (Fig. 9b). Here, the assembly relationship stored in the 3D CAD model is visualized in the tree view window on the left-hand side of the module, the 3D shape of the 3D CAD model is visualized at the center, and the nonshape data and catalog property information for the plant item selected by the user are visualized in the property window on the right-hand side of the module. These visualization-based verification experiments were used to determine whether all plant items normally referred to the catalog, and whether the information of the neutral 3D design data matched that of the native 3D design data.

As shown in Fig. 9c, the neutral 3D design data for which visualization-based verification was completed were then converted to the lightweight 3D model for use in the collaborative project management system. As a result of lightweighting, the size of the entire model file for the PDMS SNG project was reduced from 481 to 31.6 MB in conversion time of 1 h 21 s. When the same model file was lightened in the JT format, the file size was reduced to 122 MB, and the conversion time was 52 min 24 s. In addition, a lightweighting experiment with the translated neutral model data was performed on the Smart3D 3D sample, as shown in Fig. 9d. As a result of lightweighting, the size of the entire model file for the Smart3D 3D sample was reduced from 6.22 to 1.89 MB in conversion time of 4 min 35 s. When the same model file was lightened in the JT format, the file size was reduced to 3.45 MB, and the conversion time was 3 min and 29 s. The results are summarized in Table 2.

Table 2:

Lightweight file sizes and conversion times for test cases.

Lightweight file formatPDMS SNG project (neutral file: 481MB)Smart3D 3D design sample (neutral file: 6.22MB)
This studyFile size31.6 MB1.89 MB
Conversion time1 h 21 s4 m 35 s
JTFile size122 MB3.45 MB
Conversion time52 min 24 s3 min 29 s
Lightweight file formatPDMS SNG project (neutral file: 481MB)Smart3D 3D design sample (neutral file: 6.22MB)
This studyFile size31.6 MB1.89 MB
Conversion time1 h 21 s4 m 35 s
JTFile size122 MB3.45 MB
Conversion time52 min 24 s3 min 29 s
Table 2:

Lightweight file sizes and conversion times for test cases.

Lightweight file formatPDMS SNG project (neutral file: 481MB)Smart3D 3D design sample (neutral file: 6.22MB)
This studyFile size31.6 MB1.89 MB
Conversion time1 h 21 s4 m 35 s
JTFile size122 MB3.45 MB
Conversion time52 min 24 s3 min 29 s
Lightweight file formatPDMS SNG project (neutral file: 481MB)Smart3D 3D design sample (neutral file: 6.22MB)
This studyFile size31.6 MB1.89 MB
Conversion time1 h 21 s4 m 35 s
JTFile size122 MB3.45 MB
Conversion time52 min 24 s3 min 29 s

To further verify the lightweighting capability, an experiment was performed on large 3D design data in PDMS RVM format, as shown in Fig. 9e. Here, the file size was reduced from 507 to 44.3 MB for the sizable 3D design data. In addition, when the same data were converted to the JT format, the file size was reduced to 214 MB.

The completely converted lightweight 3D model was then loaded from the collaborative project management system using the VIZZARD platform (SOFTHILLS) (SOFTHILLS VIZZARD, 2020) and checked for issues when performing the functions required for collaboration support (Fig. 10). Thus, the converted lightweight 3D model was applicable to project data integration, crosssection visualization, project reviewing, dimensional checking, and interference checking without issue. Thus, by operating the collaborative project management system in web- and mobile-based environments, 3D design data can be utilized efficiently and effectively in part production, assembly, and installation sites where the use of the plant 3D CAD system is difficult.

Use of lightweight 3D model for project management support.
Figure 10:

Use of lightweight 3D model for project management support.

Since the collaborative project management system using a lightweight 3D model has small memory occupancy and data interoperability through a neutral model, it can be used in virtual reality and augmented reality environments by mounting wearable devices on it in the smart manufacturing field (Han, 2020; Rauch & Vickery, 2020). The data of the lightweight 3D model used in the collaboration management system are small in size; however, it is difficult to modify the model. Therefore, if a design change occurs, the 3D design data must first be modified in the plant 3D CAD system, converted to a lightweight 3D model again according to the proposed method, and then uploaded in the collaborative project management system.

8. Conclusions

In this paper, a 3D design data interface has been proposed to translate 3D design data created in a plant 3D CAD system according to a neutral model defined in reference to relevant international standards. In addition, after a lightweighting process, the translated 3D design data can be used in a collaborative work environment. The proposed interface primarily comprises a 3D CAD model translation unit, a catalog translation unit, a 3D design data verification unit based on a neutral model, and a conversion to lightweight 3D models unit. The 3D CAD model translation unit converts the 3D CAD model into a neutral format, and the catalog translation unit converts the equipment and material catalog into a neutral format. The 3D design data verification unit checks the consistency and accuracy of the translated neutral 3D design data. Finally, the conversion unit converts the inspected neutral 3D design data into a lightweight 3D model.

A prototype system comprising four modules was implemented to verify the proposed 3D design data interface. This prototype was used to translate and reduce the data size of 3D design data of 5225 piping components and 616 equipment units of a synthesized natural gas production plant using field data. After converting the 3D design data to a lightweight 3D model, the lightweight 3D model was loaded and employed in a collaborative project management system to confirm that there were no issues relating to the use of the model in the system’s collaboration functions.

This study has three limitations to address. First, verification of the translated neutral 3D design data is performed visually in many inspection items, including data values’ accuracy, except for verifying whether required data fields are filled and linking between 3D CAD model and catalog data is correct. Therefore, follow-up studies on verification are required. Second, disciplines supported by neutral 3D design models are limited to piping and equipment designs, and the neutral 3D design model must be extended to support other disciplines, including structural, electrical, and HVAC designs. Third, several mapping files used to translate plant 3D design data into neutral model data are currently being prepared manually. Thus, in future, it is necessary to develop a method to generate such mapping files automatically or semiautomatically.

Acknowledgments

This research was supported by the Industrial Core Technology Development Program (Project ID: 20000725&20009324), which was funded by the Ministry of Trade, Industry and Energy, Korea, and by the Basic Science Research Program (Project ID: NRF-2019R1F1A1053542) of the National Research Foundation (NRF), which was funded by the Ministry of Science and ICT (MSIT), Korea.

Conflict of interest statement

None declared.

References

Braaksma
A. J.
,
Klingenberg
W. W.
,
van Exel
P. P.
(
2011
).
A review of the use of asset information standards for collaboration in the process industry
.
Computers in Industry
,
62
(
3
),
337
350
.

Cho
J.
,
Han
S.
,
Kim
H.
(
2006
).
Meta-ontology for automated information integration of parts libraries
.
Computer-Aided Design
,
38
(
7
),
713
725
.

Eigner
M.
,
Handschuh
S.
,
Gerhardt
F.
(
2010
).
Concept to enrichen lightweight, neutral data formats with CAD-based feature technology
.
Computer-Aided Design and Applications
,
7
(
1
),
89
99
.

Fiorentini
X.
,
Paviot
T.
,
Fortineau
V.
,
Goblet
J. L.
,
Lamouri
S.
(
2013
).
Modeling nuclear power plants engineering data using ISO 15926
. In
Proceedings of 2013 International Conference on Industrial Engineering and Systems Management (IESM)
(eds
Aboutajdine
D.
,
Skalli
A.
,
Benchekroun
B.
,
Artiba
A.
) (pp.
901
906
.).
IEEE
.

Han
S.
(
2020
).
A review of smart manufacturing reference models based on the skeleton meta-model
.
Journal of Computational Design and Engineering
,
7
(
3
),
323
336
.

Irgens
T. A.
,
Hansen
T.
,
Haenisch
J.
(
2004
).
PLCS pilot for New Norwegian frigates
. In
Proceedings of the 13th Product Data Technology Europe Symposium
.

JT File Format
.
Available online: https://en.wikipedia.org/wiki/JT(visualization_format) (accessed on 1 August 2020)
.

Kim
S.
,
Lee
K.
,
Hong
T.
,
Kim
M.
,
Jung
M.
,
Song
Y.
(
2005
).
An integrated approach to realize multi-resolution of B-rep model
. In
Proceedings of the 2005 ACM Symposium on Solid and Physical Modeling
.

Kim
B. C.
,
Teijgeler
H.
,
Mun
D.
,
Han
S.
(
2011
).
Integration of distributed plant lifecycle data using ISO 15926 and Web services
.
Annals of Nuclear Energy
,
38
(
11
),
2309
2318
.

Kim
B. C.
,
Mun
D.
,
Han
S.
,
Pratt
M. J.
(
2011
).
A method to exchange procedurally represented 2D CAD model data using ISO 10303 STEP
.
Computer-Aided Design
,
43
(
12
),
1717
1728
.

Kim
B. C.
,
Mun
D.
(
2014a
).
Feature-based simplification of boundary representation models using sequential iterative volume decomposition
.
Computers & Graphics
,
38
,
97
107
.

Kim
B. C.
,
Mun
D.
(
2014b
).
Stepwise volume decomposition for the modification of B-rep models
.
The International Journal of Advanced Manufacturing Technology
,
75
(
9–12
),
1393
1403
.

Kim
B. C.
,
Mun
D.
(
2015
).
Enhanced volume decomposition minimizing overlapping volumes for the recognition of design features
.
Journal of Mechanical Science and Technology
,
29
(
12
),
5289
5298
.

Kim
B. C.
,
Jeon
Y.
,
Park
S.
,
Teijgeler
H.
,
Leal
D.
,
Mun
D.
(
2017
).
Toward standardized exchange of plant 3D CAD models using ISO 15926
.
Computer-Aided Design
,
83
,
80
95
.

Kim
B. C.
,
Kim
B.
,
Park
S.
,
Teijgeler
H.
,
Mun
D.
(
2020
).
ISO 15926–based integration of process plant life-cycle information including maintenance activity
.
Concurrent Engineering
,
28
(
1
),
58
71
.

Kim
B. C.
,
Lee
H.
,
Mun
D.
,
Han
S.
(
2021
).
Lifecycle management of component catalogs based on a neutral model to support seamless integration with plant 3D design
.
Journal of Computational Design and Engineering
,
8
(
1
),
409
427
.

Koo
S.
,
Lee
K.
(
2002
).
Wrap-around operation to make multi-resolution model of part and assembly
.
Computers & Graphics
,
26
(
5
),
687
700
.

Kwon
S.
,
Kim
B. C.
,
Mun
D.
,
Han
S.
(
2015
).
Graph-based simplification of feature-based three-dimensional computer-aided design models for preserving connectivity
.
Journal of Computing and Information Science in Engineering
,
15
(
3
),
031010
.

Kwon
S.
,
Kim
B. C.
,
Hwang
H.
,
Mun
D.
,
Han
S.
(
2016
).
Enhancement of equipment information sharing using three-dimensional computer-aided design simplification and digital catalog techniques in the plant industry
.
Concurrent Engineering
,
24
(
3
),
275
289
.

Kwon
K.
,
Mun
D.
, (
2019
).
Part recognition-based simplification of triangle mesh models for ships and plants
.
The International Journal of Advanced Manufacturing Technology
,
105
(
1–4
),
1329
1342
.

Kwon
S.
,
Mun
D.
,
Kim
B. C.
,
Han
S.
,
Suh
H. W.
(
2019
).
B-rep model simplification using selective and iterative volume decomposition to obtain finer multi-resolution models
.
Computer-Aided Design
,
112
,
23
34
.

Kwon
K.
,
Mun
D.
(
2020
).
A method to minimize the data size of a lightweight model for ship and offshore plant structure using part characteristics
.
Journal of Marine Science and Engineering
,
8
(
10
),
763
.

Leal
D.
(
2005
).
ISO 15926 Life cycle data for process plant: An overview
.
Oil & Gas Science and Technology
,
60
(
4
),
629
637
.

Lee
S.
,
Han
S.
,
Mun
D.
(
2012
).
Integrated management of facility, process, and output: data model perspective
.
Science China Information Sciences
,
55
(
5
),
994
1007
.

Li
J.
,
Kim
I.
,
Lee
S.
,
Han
S.
,
Lee
C.
,
Cheon
S.
,
Lee
W.
,
An
K.
,
Cho
G.
,
Hwang
J.
,
Mun
D.
(
2011
).
Sharing piping CAD models of ocean plants based on international standards
.
Journal of Marine Science and Technology
,
16
(
1
),
76
83
.

Nguyen
C. H. P.
,
Choi
Y
. (
2019
).
Triangle mesh and boundary representation combined approach for 3D CAD lightweight representation for collaborative product development
.
Journal of Computing and Information Science in Engineering
,
19
(
1
),
011009
.

OBJ file format
, (
accessed on 1 August 2020
).

Owen
J.
(
1997
)
STEP: An introduction
. (2nd ed.).
Information Geometers
.

Pratt
M. J.
,
Anderson
B. D.
,
Ranger
T.
(
2005
).
Towards the standardized exchange of parameterized feature-based CAD models
.
Computer-Aided Design
,
37
(
12
),
1251
1265
.

qHub
, (
accessed on 20 October 2016
).

Rauch
E.
,
Vickery
A. R.
(
2020
).
Systematic analysis of needs and requirements for the design of smart manufacturing systems in SMEs
.
Journal of Computational Design and Engineering
,
7
(
2
),
129
144
.

Safdar
M.
,
Jauhar
T. A.
,
Kim
Y.
,
Lee
H.
,
Noh
C.
,
Kim
H.
,
Lee
I.
,
Kim
I.
,
Kwon
S.
,
Han
S.
(
2020
).
Feature-based translation of CAD models with macro-parametric approach: issues of feature mapping, persistent naming, and constraint translation
.
Journal of Computational Design and Engineering
,
7
(
5
),
603
614
.

SOFTHILLS VIZZARD
,
http://www.softhills.net/vizzard.html (accessed on 9 November 2020)
.

STL file format
,
URL: https://en.wikipedia.org/wiki/STL_(file_format) (accessed on 1 August 2020)
.

Sun
R.
,
Gao
S.
,
Zhao
W.
(
2010
).
An approach to B-rep model simplification based on region suppression
.
Computers & Graphics
,
34
(
5
),
556
564
.

Woo
Y.
(
2009
).
Automatic simplification of solid models for engineering analysis independent of modeling sequences
.
Journal of Mechanical Science and Technology
,
23
(
7
),
1939
1948
.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected]