Virtual Build of Assembly Line on the 3DEXPERIENCE Platform

I.INTRODUCTION The planning of a factory is intricately linked to the product design and planning process; the life-cycle of a factory is congruent with the life-cycle of the products it is associated with. Factories are comparable to large and intricate products that require extensive and exhaustive planning during their life-cycle from the inception to the break-up stage [1]. Strategic components tend to be manufactured by the companies that produce the final products. For these strategic components firms define long term plans. As a consequence, the technological characteristics, even if in continuous evolution, can be predicted with high reliability. Less critical components tend to be externalized. In a context of continuous cost reduction, the producers of components try to obtain economies of scale by enlarging their size while specializing on specific types of components [2]. Tolio et al. explain the design of a manufacturing system is a complex task of creating manufacturing strategy decisions strictly with a view to have a long term impact [3]. This involves a significant financial outlay and the complexity in the context of generating profits on the capital investment requires a methodical approach in assessing the performance of the system. Furthermore, there is a greater demand for the flexibility of factories in order to meet continuously changing market demands, technology advances and government regulations. In this context, the concept of a virtual replica or digital twin of the manufacturing system in order to simulate and validate the decision process gains relevance. Grieves explains the concept of a digital twin of a factory and how it enables engineers to see the progress of the physical product as it is moving and actually see information about the characteristics of the physical product [4][5]. Engineers can simulate the manufacturing environment that creates the product and the operations associated with the product creation. A virtual factory must fulfill three main demands: (1) that it be able to incorporate partners at any stage of the product life-cycle, (2) that it be able to incorporate partners with varying levels of IT capabilities, and (3) that it be able to provide all required functionality [6]. Cerrone et al. performed tests to determine the crack path of a specimen by using as-manufactured models [7]. Although they utilized only a few functionalities of the digital twin, they demonstrated the high degree of fidelity of a digital twin to the physical product. In this paper, the steps undertaken to create a 3-dimensional virtual factory from a 2-dimensional draft are described in the context of a motorcycle frame assembly line.   II.THE 3DEXPERIENCE PLATFORM   The 3DEXPERIENCE platform is a BUSINESS EXPERIENCE platform offered by DASSAULT SYSTEMES that provides software solutions for every organization in a company - from marketing to sales to engineering – that help in the value creation process, to create differentiating consumer experiences. The 3DEXPERIENCE platform allows enterprises to connect data and people through an interactive and user-friendly interface, both inside and outside the enterprises’ domains. The interface provided contains applications related to people collaboration, modelling and simulation. The proprietary workbenches of DASSAULT SYSTEMES have been grouped under a broad classification of apps, namely Social and Collaborative Apps, Information Intelligence Apps, 3D Modelling Apps and Simulation Apps.     III.Methodology   Engineering applications can be run on smartphones or tablets as part of mobile solutions to enable workers to make decisions on site, as well as on conventional devices such as personal computers. Engineering apps provide a single collaborative environment for industrial, manufacturing engineers, and management personnel to develop and provide the best manufacturing flow practices throughout the production design process [8][9]. The 3DEXPERIENCE platform used throughout the creation of the virtual line has both on-premises and cloud computing capabilities. A key feature that has been exploited is the ability to work on large data by using high performance computers and being able to share the data through a cloud-based environment with partners. Bzymek et al. have simulated a machining sequence and demonstrated the scope of DELMIA in integrating not only manufacturing optimization, but also design and planning capabilities as well as line balancing [10]. The Plant Layout Design application featured in the DELMIA workbench was used throughout the Resource Structuring of the Frame Assembly Line.  

A.Creating the Footprint

The draft of the Frame Assembly Line consisted of detailed 2-dimensional representations of every resource associated with it. In order to create a useable drawing of the frame assembly line, the representations of the resources were simplified to resemble line and box elements as illustrated in Fig.1. A drafting software can be used to perform this operation. The simplification of the drawing is necessary to avoid alignment errors caused by ambiguities in the 2-dimensional representation of certain elements while structuring the 3-dimensional environment.      

B.Importing the Footprint into 3DEXPERIENCE

A Product Process Resource (PPR) context was created in the Plant Layout Design application to represent the plant. Within the PPR context, a Manufacturing Cell and its associated Area resource were created. The Area resource provides a virtual space for the creation of a footprint representation. The Frame Assembly Line footprint was inserted onto the representation created in the Area resource. The required footprint is automatically converted into the format associated with the 3DEXPERIENCE platform while the file is imported into the database. Modifications to the inserted footprint such as the addition of text in the form of labels were made using Footprint Edition.  

C.Generating Resources

The resources to be placed on the footprint were created as 3-dimensional physical products. The 3-dimensional virtual representations of the resources were saved as templates and added to a Resource Catalog. The resource type is defined at the time of creation of the template. Resources can be inserted into the PPR context by copying instances of the physical product under the desired node in the resource tree, or by generating them from a catalog. Fig-2 shows the virtual representation of a conveyor that has been generated in the digital work space.      

D.Resource Allocation

The footprint contains demarcations specifying the resource locations for each resource associated with the Frame Assembly Line. The height and orientation of resources such as conveyors, jigs, hangers, and floor level machines are given in SI units at the resources’ locations. Resources were snapped onto the footprint at their respective locations and their absolute position in space was controlled by using the robot in the Plant Layout Design application. The robot is flexible and can be snapped onto any surface or point, and at any orientation on the resource. Resources can be snapped to each other or to the footprint. Multiple instances of resources were generated as patterns and each pattern was aligned according to a specific resource, or distributed along a defined axis. The outcome of the resource allocation can be seen in Fig.3 as the interaction between multiple resource types and instances.  

E.Validating the virtual build

The Bill of Materials associated with the product to be manufactured is defined and the Process Planning for the product assembly is done using a top-down approach. The resource assignment is performed at the process planning stage by using a library of resources in common with the virtual construction of the assembly line. The assembly of the product is evaluated by defining the motion and the time taken for each component to be assembled into the complete product. In order to validate the virtual build, the assembly process is reviewed in the context of the 3D environment created in the previous steps. In the scenario considered, the assembly of the motorcycle frame involves multiple components sourced from various locations of the factory. The worker walkways are to be minimized while performing the assembly with the minimum of effort and time. This can be achieved by validating the positioning and orientation of the assembly steps by combining the data involved with the virtual factory dataset. The virtual build is flexible in allowing modifications to be made in the process plan and the assembly steps. After each modification, the assembly simulation is performed to optimize the chain of operations. The worker instructions are defined after obtaining the optimized manufacturing solution.   IV.NEED FOR A VIRTUAL FACTORY   Before the Industry 4.0 or fourth industrial revolution, the only way to gain detailed information about the status of any operating industrial equipment was to be in physical proximity to it and inspect it. In spite of incorporating lean manufacturing principles, time and human effort are still expended in order to create an optimal solution. Digital manufacturing reduces the disadvantages associated with the implementation of lean manufacturing principles, and fulfills the need to reduce unnecessary elements of the process while allowing theoretically infinite ways of fine-tuning the scenario under consideration. A virtual build is the first stage of developing a capable and detailed digital manufacturing solution. Industries have already started implementing the concept of a virtual build as an entry point to digital manufacturing solutions. CAD 3D models, for example, are rich and accurate digital representations that allow designers to determine how different parts fit together. Manufacturing simulations can also determine whether virtual designs can actually be built using the machines available. The exact state and condition of a physical object can also be determined anywhere across the world through the use of real-time data feeds from sensors. The real advantage of the virtual model lies in the ability to synchronize every aspect from design to real-time data feed in order to optimize the lifetime of the factory assets. A virtual build is crucial in reducing construction and prototyping costs, while combining real-time data with an accurate digital model of the physical product allows for the prediction of failure, thus reducing maintenance costs and downtime. A particular benefit in the case of large manufacturing industries with multiple units is the ability to obtain and analyze data from each unit to create the best solution. The development of methods and techniques to disseminate the knowledge is gaining in importance because a company can improve its performance if it is capable of generating, absorbing, and sharing knowledge and good practices. This could help increase productivity, reduce processing times and re-work and improve product quality, and also optimize the exploitation of industrial resources, improve the communication among different teams and promote innovation inside the company.   V.CONCLUSION   The Factory Level comprises of product design, process development, production planning, and operations. A virtual factory enables the various stages of the system chain to be streamlined through the use of identical datasets, eliminating the ambiguity caused by different human perception. Through the use of an inter-connected environment such as the 3DEXPERIENCE platform, industries can connect across the globe and share data through both public and private clouds. Powerful on-premises capabilities of such a platform also enable true-to-life representations of the elements of the digital factory. The creation of the 3-dimensional model of the motorcycle frame assembly line in the scenario considered in this paper is achieved by a standard process that can be replicated on varying computer systems, thus ensuring a uniform dataset and a collaborative environment across the system chain.   References

1. Azevedo.A, Almeida.A, “Factory Templates for Digital Factories Framework”, Robotics and Computer-Integrated Manufacturing, Elsevier, 27 (2011), 755–771
2. W. Terkaj, T. Tolio, A. Valente, “Designing Manufacturing Flexibility in Dynamic Production Contexts”, Design of Flexible Production Systems, Chapter: 7(pp.137-190), Springer Berlin Heidelberg,
3. T. Tolio, M. Saccob, W. Terkajb, M. Urgoa, “Virtual Factory: an Integrated Framework for Manufacturing Systems Design and Analysis”, Procedia CIRP 7 ( 2013 ) 413 – 418
4. M.W. Grieves, ”Digital Twin: Manufacturing Excellence through Virtual Factory Replication”, Digital Twin White Paper, LLC 2014
5. M. Grieves, J. Vickers, “Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems” (Excerpt) from Trans-Disciplinary Perspectives on System Complexity
6. D.M. Upton, A. McAfee (1996, July-August), “The Real Virtual Factory”, Harvard Business Review[Online] https://hbr.org/1996/07/the-real-virtual-factory.
7. A. Cerrone, J. Hochhalter, G. Heber, A. Ingraffea, “On the Effects of Modeling As-Manufactured Geometry: Toward Digital Twin”, International Journal of Aerospace Engineering Volume 2014
8. M. Neumanna, E. Westkämper, “Method for situation-based Modeling and Simulation of Assembly Systems”, Procedia CIRP 7 ( 2013 ) 413 – 418
9. J. W. Volkmanna, M. Landherra, D. Luckea, M. Saccob, M. Lickefetta, E. Westkämper “Engineering apps for advanced industrial engineering”, Procedia CIRP 41 ( 2016 ) 632 – 637
10. Z.M. Bzymek, M. Nunez, M. Li and S. Powers, “Simulation of a Machining Sequence Using Delmia/Quest Software”, Computer-Aided Design and Applications, 5(1-4), 2008, 401-411
11. A. Cerrone, J. Hochhalter,  G. Heber, A. Ingraffea, “On the Effects of Modeling As-Manufactured Geometry: Toward Digital Twin”, International Journal of Aerospace Engineering Volume 2014