- Email: [email protected]
Digital Twin of a Cutting Tool
Available online atonline www.sciencedirect.com Available online at Available at www.sciencedirect.com Available online at www.sciencedirect.com www.sciencedirect.com
ScienceDirect ScienceDirect Procedia CIRP 00 (2017) 000–000 Procedia CIRP(2018) 72 (2018) 215–218 Procedia Procedia CIRP CIRP 00 00 (2018) 000–000 000–000
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
51st 51st CIRP CIRP Conference Conference on on Manufacturing Manufacturing Systems Systems
Digital of Cutting Tool Digital Twin of aa May Cutting Tool France 28th CIRP DesignTwin Conference, 2018, Nantes, a,* b c a a Darya Darya Botkina Botkinaa,* ,, Mikael Mikael Hedlind Hedlindb ,, Bengt Bengt Olsson Olssonc ,, Jannik Jannik Henser Hensera ,, Thomas Thomas Lundholm Lundholma
A new methodology to ofanalyze the functional and physical architecture of KTH Technology, KTH Royal Royal Institute Institute of Technology, PMH PMH Application Application Lab, Lab, Brinellv¨ Brinellv¨a agen gen 68, 68, Stockholm Stockholm 10044, 10044, Sweden Sweden Scania 87 Scania CV CV AB, AB, 151 151 87 S¨ S¨o odert¨ dert¨a alje, lje, Sweden Sweden existing products for an assembly oriented product family identification Sandvik Coromant, Mossv¨ a gen 10, 811 81 Sandvik Coromant, Mossv¨agen 10, 811 81 Sandviken, Sandviken, Sweden Sweden a a
c c
∗ ∗
b b
Corresponding Corresponding author. author. Tel.: Tel.: +46 +46 8 8 790 790 6339. 6339. E-mail E-mail address: address: [email protected] [email protected]
Abstract Abstract
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
*This Corresponding author. +33 3twin 87 37 30; E-mail [email protected] paper focuses onTel.: a digital of54 a cutting tooladdress: as a digital replica of a physical tool, its data format and structure, information flows and data
This paper focuses on a digital twin of a cutting tool as a digital replica of a physical tool, its data format and structure, information flows and data management, management, as as well well as as possibilities possibilities for for further further applications applications and and analysis analysis of of productivity. productivity. Data Data are are collected collected throughout throughout the the production production lifecycle lifecycle in an accurate way, using the international standard ISO 13399 and messaging based on the previously developed event-driven in an accurate way, using the international standard ISO 13399 and messaging based on the previously developed event-driven line line information information system system architecture architecture (LISA) (LISA) with with IoT IoT functionality. functionality. The The digital digital twin twin is is tweeted tweeted to to be be stored, stored, refined refined and and propagated propagated to to the the process process planning planning for for an an Abstract optimized optimized machining machining solution. solution. cc 2018 Authors. Published by by Elsevier Elsevier B.V. 2018 The The Authors. B.V. more product variety and customization is unbroken. Due to this development, the need of © Published In today’s business environment, thethe trend towards Peer-review under responsibility of scientific committee of CIRP Conference on Manufacturing Systems. Peer-review underresponsibility responsibility ofthe the scientific committee ofthe the51st 51st CIRPproducts Conference on Manufacturing Systems. Peer-review under of scientific committee of the 51st CIRP Conference Manufacturing agile and reconfigurable production systems emerged to cope with various andon product families. Systems. To design and optimize production
systems as well astwin; to choose product product analysis methods are needed. Indeed, most of the known methods aim to digital cutting tool; internet things; manufacturing information system Keywords: Keywords: digital twin; cutting the tool;optimal internet of of things; matches, manufacturing information system analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and nature of components. This fact impedes an efficient comparison and choice of appropriate product family combinations for the production system. A new methodology is proposed to analyze existing products in view of their functional and physical architecture. The aim is to cluster flow of analysis, which in turn results in 1. flow of data data facilitates facilitates analysis, in of turn results in aa better better 1. Introduction Introduction these products in new assembly oriented product families for the optimization of existing assembly lines and thewhich creation future reconfigurable production outcome. production outcome. assembly systems. Based on Datum Flow Chain, the physical structure of the products is analyzed. Functional subassemblies are identified, and The access data of gives the The digitalization of opens a functional analysis is performed. Moreover, a hybrid functional and physical graph (HyFPAG) is the outputstep which depicts the Thearchitecture access to to the the data of every every production production step gives the proproThe ever-increasing ever-increasing digitalization of manufacturing manufacturing opens duction system a previously unknown flexibility and allows up multiple opportunities to significantly increase productivity similarity between product families by providing design support to both,duction production system planners and product flexibility designers. An system a previously unknown andillustrative allows it it up multiple opportunities to significantly increase productivity respond efficiently human control. This examand To aa proper of aa complex example of a nail-clipper is used to explain behaviour the proposed industrial study onwithout two product families of steering columns of to respondcase efficiently without human control. This paper paper examand effectiveness. effectiveness. To ensure ensure proper behaviour ofmethodology. complex Anto ines of twin production modern engineering model based simthyssenkrupp Presta France is then carried outuses to give a first industrial of concept the proposed approach. ines the the concept of digital digital twin for for cutting cutting tools tools and and production production production system system modern engineering uses model based sim- evaluation ©ulations 2017 The Authors. Published by Elsevier B.V. systems as a powerful method to represent an and data analytics at all stages, not only at the initial systems as a powerful method to represent an industrial industrial proproulations and data analytics at all stages, not only at the initial Peer-review under theproduction scientific committee the 28th Design Conference cess and its parts, in 2018. other words, a bridge between ‘real’ and design stage, butresponsibility also duringofthe process of itself, to CIRP
design stage, but also during the production process itself, to predict predict the the outcome, outcome, optimize, optimize, correct correct and and evaluate. evaluate. Such Such ononKeywords: Assembly; Design method; Family identification going data collection, simulations or virtual production going data collection, simulations or virtual production models, models, necessitated necessitated by by requirements requirements of of the the modern modern competitive competitive indusindustrial environment, are usually called “digital trial environment, are usually called “digital twins” twins” [1,3,5–7]. [1,3,5–7]. Digitalization Digitalization of of planning planning and and development development of of the the modern modern 1. Introduction production process allows to do away with the expensive production process allows to do away with the expensive physphysical mockups of eventual production line: aa digital twin ical mockups of the thefast eventual production in line:the digital twin enenDue to the development domain of ables development decision ables aa cost-efficient cost-efficient process development and decision making making communication and process an ongoing trend and of digitization and at or stage. at the the operational operational or servicing servicingenterprises stage. digitalization, manufacturing are facing important Modern production lines work with flows. Modern in production workenvironments: with massive massive data data flows. A A challenges today’s lines market a continuing digital twin can account for the ever increasing need for digital twin can account forofthe ever increasing needtimes for data data tendency towards reduction product development and accumulation and data at those of accumulation and lifecycles. data processing processing at all all stages: stages: those of design, design, shortened product In addition, there is an increasing development, testing, actual production, assessment and development, testing, actualbeing production, assessment anda correccorrecdemand of customization, at the same time in global tion of To aa digital twin, data tion or or adjustment adjustment of the the process. process. To create create digitalThis twin,trend, data competition with competitors all over the world. from various sources are collected and analyzed, such as physfrom various sourcesthe are development collected and analyzed, suchtoas micro physwhich is inducing from macro ical manufacturing information, data ical dimensions, dimensions, manufacturing information, operational data markets, results in diminished lot sizes due operational to augmenting and information flows from analytic software. This informaand information flows from analytic software.production) This informaproduct varieties (high-volume to low-volume [1]. tion is into aa virtual model. designed approtioncope is combined combined virtualvariety model. asWhen When approTo with this into augmenting well designed as to be able to priately, model results accurate simulation of priately, this this modeloptimization results in in aa very very accurate in simulation of the the identify possible potentials the existing actual assets and their operandi. The ongoing actual physical physical assets their modus modus operandi. ongoing production system, it isand important to have a preciseThe knowledge
cess and its parts, in other words, a bridge between ‘real’ and ‘digital’ ‘digital’ world. world.
2. 2. ISO ISO 13399 13399 –– cutting cutting tool tool data data representation representation and and exexchange change of the product range and characteristics manufactured and/or assembled in this system. In this context, the main challenge in ISO the international for ISO 13399 13399 isanalysis the current current international technical standard for modelling and is is now not onlytechnical to cope standard with single information exchange between various software servicing cutinformation exchange between various software servicing cutproducts, a limited product range or existing product families, ting tools and holders, as manufacturers tingalso tools andable holders, as well well astobetween between tool manufacturers but to be to analyze andas comparetool products to define and suppliers. It is developed to facilitate the use, manipulation and suppliers. It is developed facilitate that the use, manipulation new product families. It can betoobserved classical existing and of cutting tool within among and exchange exchange ofare cutting tool data data within and and amongormanufacmanufacproduct families regrouped in function of clients features. turing, and turing, distribution, distribution, and usage. usage. However, assembly oriented product families are hardly to find. When the operation of tools based aa standardized forWhen operation of level, tools is isproducts based on ondiffer standardized forOn the the product family mainly in two mat, the information conversion quality and speed will be sigmat, the information (i) conversion quality and speed will be sigmain characteristics: the number of components and (ii) the nificantly improved, and the run In nificantly improved,(e.g. and mechanical, the system system will will run smoothly. smoothly. In the the type of components electrical, electronical). long run ISO boosts exchange efficiency long run the the methodologies ISO 13399 13399 standard standard boosts the the exchange Classical considering mainly singleefficiency products and the quality. At time, it andsolitary, the communication communication quality.product At the the same same time, analyze it essentially essentially or already existing families the liberates the user from any additional work needed to liberatesstructure the useron from any additional work needed to initiate initiate product a physical level (components level) which the exchange various parts of assembly the data data difficulties exchange between between various of the the definition assembly [1]. [1]. causes regarding an parts efficient and ISO 13399 defines the information structure needed deISO 13399 defines the information structure needed to tothis decomparison of different product families. Addressing scribe scribe various various data data about about cutting cutting tool tool assemblies. assemblies. The The standard standard
cc© 2018 2212-8271 The Authors. Published by Elsevier B.V. 2212-8271© 2018The The Authors. Published by Elsevier 2212-8271 2018 The Authors. Published byElsevier Elsevier B.V. B.V. 2212-8271 2017 Authors. Published by B.V. Peer-review under responsibility of the scientific committee of 51st on Systems. Peer-review under responsibility of scientific the scientific committee theCIRP 51stConference CIRP Conference Conference on Manufacturing Peer-review under responsibility of the the scientific of the the of 51st CIRP Conference on Manufacturing Manufacturing Systems. Systems. Peer-review under responsibility of committee 28th CIRP Design 2018. 10.1016/j.procir.2018.03.178
216 2
Darya Botkina et al. / Procedia CIRP 72 (2018) 215–218 D. Botkina et al / Procedia CIRP 00 (2018) 000–000
helps to offer a safe and convenient communication environment for [4]:
about the new assembled tool can be send by using internet of things modularity [1].
• Tool item: version, definition, properties, mutual relationship between each other etc. • Tool assembly: definition, properties, assembled instructions, additional data for 3D model software (CAD, CAM/CNC). • References to external documents: 3D model of a single or tool assembly. • Multifunctionality: tools being used for more than one purpose. • Nominal and physical tools: nominal tool information to CAM and the tool room. Besides, physical tool information between the tool room and CNC. The data about the tool and the tool assembly are broken into the following four major classes, or modules, and Fig. 1 represents an assembly structure of cutting tools [2]. • Cutting item: information about the cutting edge of a tool, i.e. inserts and heads of solid tools. • Tool item: information about tool holders of cutting items. i.e. solid bar or solid head which is used for holding inserts in place. • Adaptive item: information about adapters, i.e. adaptive connections between a cutter and a machine. • Assembly item: information about small joints, i.e. components that join different parts of the assembly together (screws, clamps, etc).
Fig. 2. Concept evaluation, CAM simulation and ToolMaker application environment
In this way, cutting tool data are structured and standardized, which facilitates the connection between all parties of production, and creates 3D models for further applications (see Fig. 2). 3. Information flow 3.1. LISA – line information system architecture
Fig. 1. Different type of items and assembly [2]
This data structure largely resolves the problem of tool classification, for instance, two different tools can be made up from the same “catalog items”, but set to different nominal dimensions, thus there is no danger of confusion at the simulation stage. The verification and implementation of the standard ISO 13399 is provided by a software called ToolMaker. It was developed by Sandvik Coromant initially as an open source application for the practical use of the standard, and currently is updated by KTH Royal Institute of Technology, adding new modules and functions. ToolMaker provides a convenient and efficient interface for the representation of tool information in a structured way (Fig. 2). Users can create a library of necessary items and tools; these files can be used to build an assembly. The assembly can be exported as 3D model and property file. In addition, information
Abundance of communication protocols and formats make information exchange between different levels and stages of production more difficult. For example, many companies prefer to use the point-to-point communication format which necessitates an update of the entire system to include a new application [8]. KTH Royal Institute of Technology initiated the development of a flexible information architecture, called LISA (line information system architecture), that would facilitate communication among equipment of different origin, capabilities or ages. LISA is an event-based service oriented architecture (SOA) [8]. LISA allows individual applications to render themselves as services in the architecture so that other application would be able to interface with them. LISA results in an architecture which keeps applications loosely attached and retains a flexible message structure for an efficient integration of applications. The central elements of LISA are the message formats, communication and service endpoints and the message bus. The message bus, referred to as the enterprise service bus (ESB), enables LISA to transform events into usable data in a sufficiently
Darya Botkina et al. / Procedia CIRP 72 (2018) 215–218 D. Botkina et al / Procedia CIRP 00 (2018) 000–000
217 3
Fig. 3. Demonstrator: Digital twin of a cutting tool
flexible way. An event-based architecture like LISA can manage structural changes in production, simultaneously allowing for diverse applications [8]. 3.2. Tweeting machine Communication between the production line and other possible consumers of information (e.g. process owner, CAM engineers, etc) is implemented in the form of detailed and specific messages called tweets via the so called tweeting machine. The tweeting machine allows to get real-time data from CNC machining by sending tweets. Tweets are published to the enterprise service bus (ESB) by using LISA architecture. These data, transferred to LISA, provide users with the possibility to subscribe to the information. It offers useful and structured real-time data from CNC machining, which can be utilized e.g. to aid in analyzing and decreasing processing flaws, increasing quality, decision making. The Tweeting machine project was coordinated by KTH Royal Institute of Technology and aimed to demonstrate internet of things (IoT) functionality and realize traceability in two manufacturing scenarios: 1) Manufacturing (CNC machining) tracing based on STEPNC. For example, data about the tool being used, its ongoing machining operation, toolpath, forces and torque, time of usage, etc. 2) Quality (product characteristic measurement) tracing based on STEP-NC. For example, axis-positions and data from sensors, using design criteria and tolerances. Thus, the tweeting machine is a powerful tool to get updated data at every moment of production, which can be triggered by the user or other applications continuously. Other applications have the possibility to handle these messages and process the data, update the models and analyze the data. 4. Use case Manufacturing of products for heavy vehicles, for instance, a cylinder head, usually includes the participation of a few par-
ties, as manufacturer needs adapted tool for the production. The companies have to have an efficient method for exchange of information in a way that promotes design and production. The scenario underlying the digital pilot is described below and schematically represented in Fig. 3. 1) An advanced product is designed using CAD, and proceeded to manufacture it using CAM. Suppose that at either of these stages, CAM or CAD, it turns out that a new tool from the tool provider is required for a particular operation. 2) The digital model of a tool is realized based on ISO 13399 and ToolMaker application, using the tweeting machine functionality. The initial values for the process are determined. Extra conditions, for example, available power, forces and torque, machine characteristics and any desired tool change are considered. A reference to the digital “as designed” or “nominal tool” model is tweeted to the database. 3) With product design from CAD, process description from CAM and tool description “as designed” or “nominal tool” from ToolMaker or other tool configuration utility, the machining solution can be compiled, simulated and verified. 4) The tool products are ordered and delivered. The physical tool assembly is built based on the description presented in step 2). The actual dimensions of the finished assembly are measured and tweeted, so that an instance “as realized” or “physical tool” of the digital model can be created. A reference to the digital “as realized” or “physical tool” model is tweeted to the database. 5) At the stage of part manufacturing, all data relevant to the “as used” tool are tweeted. These include measurements of parts, tool compensations and changes, cutting data, etc all with STEP-NC contexts. Here, the previously developed CNC/PLC interface publisher, STEP viewer and receiver applications are utilized for further functionality. An XML-based interface to Productivity Analyzer (Sandvik Coromant’s internal data collection system for customers), is being developed. 6) Analysis of collected data might be implemented with ma-
218 4
Darya Botkina et al. / Procedia CIRP 72 (2018) 215–218 D. Botkina et al / Procedia CIRP 00 (2018) 000–000
chine learning technology and data analytics. 7) All relevant information about tool utilization is visualized for operators. 8) Suppose that a bottle neck operation has been identified and it has been determined that a 15% reduction of its process time is needed to reaching the productivity goal of a manufacturing system. This may be due to an unpredictable tool life, or, for example, because the production personnel decides to to adjust the cutting data to keep measurements within tolerances. In this eventuality, one or more tool vendors are contacted by tweeting a request to help to increase productivity of this operation by 15%. 9) Suppose that a tool provider returns by tweeting a proposal for an optimized process or tool. At this stage, software tools can be used by the company’s specialists. 10) The optimized process is verified and introduced into production. The proposed scheme allows to realize the data flow at every step of production, and ensures the possibility for continuous improvement of process and tool. The following analysis of the data enables a better prediction of behavior, simulations and performance calculation. 5. Results and conclusion The increasing need for an effective data exchange and communication is one of the challenges in manufacturing today. The need of the industry to communicate, share and exchange information is the driving force behind the rapid digitalization of the production line. Engineers require a rapid and wellstructured access to the data collected throughout the whole production lifecycle. A digital twin is a modern solution to these ever-growing needs of digitalization. A digital twin collects data and monitors the process, has access to past data, and, overall, allows for a better understanding of the production process and better prediction of the behavior and results. In this paper, the digital twin of a cutting tool is described. It is based on the international standard for cutting tool data representation and exchange ISO 13399 and the line information system architecture LISA. An information model which collects standardized data automatically, and allows for a continuous adjustment of the digital model has been developed. The adjustment is done by means of a tweeting machine which feeds the data from the production line and helps the evolution of the digital tool from the instance “as designed”, through the instance “as realized”, eventually, to “as used”. The adjusted model accurately represents properties of the cutting tool, and is applicable to precise process simulation, control and analysis, eventually leading to the continuous improvement of the production process. We would like to emphasize two novel paradigms of the proposed solution. First, data collection and flow is based on an international standard ISO 13399 and is maintained during the whole production cycle and tool life. Second, it uses open source applications such as LISA, tweeting machine, ToolMaker, thus opening possibilities for further development. At the same time implementation of the proposed solution poses some challenges which, nonetheless are can not be con-
sidered prohibitive. For example, modern product lines do not typically have the tweeting machine capability: connectivity through tweeting requires modification of existing controllers and PLC-software. 6. Acknowledgements The presented research work is financed by VINNOVA and Powertrain Manufacturing for Heavy Vehicles Application Lab - a Collaboration between KTH, Fraunhofer and RISE. References [1] D. Botkina, M. Xu ISO ToolMaker application development, thesis, KTH, 2016. c ISO 2006, [2] Cutting tool data representation and exchange, report, [s.l.]: 2006. [3] Y. Li, M. Hedlind, T. Kjellberg, G. Sivard, Cutting Tool Data Representation and Implementation Based on STEP AP242, Smart Product Engineering, Lecture Notes in Production Engineering. Springer, Berlin, Heidelberg, 2013. [4] B. Olsson, Communicating Cutting Tool Data Using ISO13399, report, International STEP-NC Demonstration, Bath, UK 2009. [5] R. Rosen, G. Wichert, G. Lo, K. Bettenhausen, About The Importance of Autonomy and Digital Twins for the Future of Manufacturing, IFACPapersOnLine, 48:3, 2015, pp 567-572. [6] B. Schleich, N. Anwer, L. Mathieu, S. Wartzacka, Shaping the digital twin for design and production engineering, CIRP Annals, 66:1, 2017, pp 141144. [7] F. Tao, J. Cheng, Q. Qi et al., Digital twin-driven product design, manufacturing and service with big data, Int J Adv Manuf Technol (2017), https://doi.org/10.1007/s00170-017-0233-1. [8] A. Theorin et al., An Event-Driven Manufacturing Information System Architecture for Industry 4.0, International Journal of Production Research, 2016.
ScienceDirect ScienceDirect Procedia CIRP 00 (2017) 000–000 Procedia CIRP(2018) 72 (2018) 215–218 Procedia Procedia CIRP CIRP 00 00 (2018) 000–000 000–000
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
51st 51st CIRP CIRP Conference Conference on on Manufacturing Manufacturing Systems Systems
Digital of Cutting Tool Digital Twin of aa May Cutting Tool France 28th CIRP DesignTwin Conference, 2018, Nantes, a,* b c a a Darya Darya Botkina Botkinaa,* ,, Mikael Mikael Hedlind Hedlindb ,, Bengt Bengt Olsson Olssonc ,, Jannik Jannik Henser Hensera ,, Thomas Thomas Lundholm Lundholma
A new methodology to ofanalyze the functional and physical architecture of KTH Technology, KTH Royal Royal Institute Institute of Technology, PMH PMH Application Application Lab, Lab, Brinellv¨ Brinellv¨a agen gen 68, 68, Stockholm Stockholm 10044, 10044, Sweden Sweden Scania 87 Scania CV CV AB, AB, 151 151 87 S¨ S¨o odert¨ dert¨a alje, lje, Sweden Sweden existing products for an assembly oriented product family identification Sandvik Coromant, Mossv¨ a gen 10, 811 81 Sandvik Coromant, Mossv¨agen 10, 811 81 Sandviken, Sandviken, Sweden Sweden a a
c c
∗ ∗
b b
Corresponding Corresponding author. author. Tel.: Tel.: +46 +46 8 8 790 790 6339. 6339. E-mail E-mail address: address: [email protected] [email protected]
Abstract Abstract
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
*This Corresponding author. +33 3twin 87 37 30; E-mail [email protected] paper focuses onTel.: a digital of54 a cutting tooladdress: as a digital replica of a physical tool, its data format and structure, information flows and data
This paper focuses on a digital twin of a cutting tool as a digital replica of a physical tool, its data format and structure, information flows and data management, management, as as well well as as possibilities possibilities for for further further applications applications and and analysis analysis of of productivity. productivity. Data Data are are collected collected throughout throughout the the production production lifecycle lifecycle in an accurate way, using the international standard ISO 13399 and messaging based on the previously developed event-driven in an accurate way, using the international standard ISO 13399 and messaging based on the previously developed event-driven line line information information system system architecture architecture (LISA) (LISA) with with IoT IoT functionality. functionality. The The digital digital twin twin is is tweeted tweeted to to be be stored, stored, refined refined and and propagated propagated to to the the process process planning planning for for an an Abstract optimized optimized machining machining solution. solution. cc 2018 Authors. Published by by Elsevier Elsevier B.V. 2018 The The Authors. B.V. more product variety and customization is unbroken. Due to this development, the need of © Published In today’s business environment, thethe trend towards Peer-review under responsibility of scientific committee of CIRP Conference on Manufacturing Systems. Peer-review underresponsibility responsibility ofthe the scientific committee ofthe the51st 51st CIRPproducts Conference on Manufacturing Systems. Peer-review under of scientific committee of the 51st CIRP Conference Manufacturing agile and reconfigurable production systems emerged to cope with various andon product families. Systems. To design and optimize production
systems as well astwin; to choose product product analysis methods are needed. Indeed, most of the known methods aim to digital cutting tool; internet things; manufacturing information system Keywords: Keywords: digital twin; cutting the tool;optimal internet of of things; matches, manufacturing information system analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and nature of components. This fact impedes an efficient comparison and choice of appropriate product family combinations for the production system. A new methodology is proposed to analyze existing products in view of their functional and physical architecture. The aim is to cluster flow of analysis, which in turn results in 1. flow of data data facilitates facilitates analysis, in of turn results in aa better better 1. Introduction Introduction these products in new assembly oriented product families for the optimization of existing assembly lines and thewhich creation future reconfigurable production outcome. production outcome. assembly systems. Based on Datum Flow Chain, the physical structure of the products is analyzed. Functional subassemblies are identified, and The access data of gives the The digitalization of opens a functional analysis is performed. Moreover, a hybrid functional and physical graph (HyFPAG) is the outputstep which depicts the Thearchitecture access to to the the data of every every production production step gives the proproThe ever-increasing ever-increasing digitalization of manufacturing manufacturing opens duction system a previously unknown flexibility and allows up multiple opportunities to significantly increase productivity similarity between product families by providing design support to both,duction production system planners and product flexibility designers. An system a previously unknown andillustrative allows it it up multiple opportunities to significantly increase productivity respond efficiently human control. This examand To aa proper of aa complex example of a nail-clipper is used to explain behaviour the proposed industrial study onwithout two product families of steering columns of to respondcase efficiently without human control. This paper paper examand effectiveness. effectiveness. To ensure ensure proper behaviour ofmethodology. complex Anto ines of twin production modern engineering model based simthyssenkrupp Presta France is then carried outuses to give a first industrial of concept the proposed approach. ines the the concept of digital digital twin for for cutting cutting tools tools and and production production production system system modern engineering uses model based sim- evaluation ©ulations 2017 The Authors. Published by Elsevier B.V. systems as a powerful method to represent an and data analytics at all stages, not only at the initial systems as a powerful method to represent an industrial industrial proproulations and data analytics at all stages, not only at the initial Peer-review under theproduction scientific committee the 28th Design Conference cess and its parts, in 2018. other words, a bridge between ‘real’ and design stage, butresponsibility also duringofthe process of itself, to CIRP
design stage, but also during the production process itself, to predict predict the the outcome, outcome, optimize, optimize, correct correct and and evaluate. evaluate. Such Such ononKeywords: Assembly; Design method; Family identification going data collection, simulations or virtual production going data collection, simulations or virtual production models, models, necessitated necessitated by by requirements requirements of of the the modern modern competitive competitive indusindustrial environment, are usually called “digital trial environment, are usually called “digital twins” twins” [1,3,5–7]. [1,3,5–7]. Digitalization Digitalization of of planning planning and and development development of of the the modern modern 1. Introduction production process allows to do away with the expensive production process allows to do away with the expensive physphysical mockups of eventual production line: aa digital twin ical mockups of the thefast eventual production in line:the digital twin enenDue to the development domain of ables development decision ables aa cost-efficient cost-efficient process development and decision making making communication and process an ongoing trend and of digitization and at or stage. at the the operational operational or servicing servicingenterprises stage. digitalization, manufacturing are facing important Modern production lines work with flows. Modern in production workenvironments: with massive massive data data flows. A A challenges today’s lines market a continuing digital twin can account for the ever increasing need for digital twin can account forofthe ever increasing needtimes for data data tendency towards reduction product development and accumulation and data at those of accumulation and lifecycles. data processing processing at all all stages: stages: those of design, design, shortened product In addition, there is an increasing development, testing, actual production, assessment and development, testing, actualbeing production, assessment anda correccorrecdemand of customization, at the same time in global tion of To aa digital twin, data tion or or adjustment adjustment of the the process. process. To create create digitalThis twin,trend, data competition with competitors all over the world. from various sources are collected and analyzed, such as physfrom various sourcesthe are development collected and analyzed, suchtoas micro physwhich is inducing from macro ical manufacturing information, data ical dimensions, dimensions, manufacturing information, operational data markets, results in diminished lot sizes due operational to augmenting and information flows from analytic software. This informaand information flows from analytic software.production) This informaproduct varieties (high-volume to low-volume [1]. tion is into aa virtual model. designed approtioncope is combined combined virtualvariety model. asWhen When approTo with this into augmenting well designed as to be able to priately, model results accurate simulation of priately, this this modeloptimization results in in aa very very accurate in simulation of the the identify possible potentials the existing actual assets and their operandi. The ongoing actual physical physical assets their modus modus operandi. ongoing production system, it isand important to have a preciseThe knowledge
cess and its parts, in other words, a bridge between ‘real’ and ‘digital’ ‘digital’ world. world.
2. 2. ISO ISO 13399 13399 –– cutting cutting tool tool data data representation representation and and exexchange change of the product range and characteristics manufactured and/or assembled in this system. In this context, the main challenge in ISO the international for ISO 13399 13399 isanalysis the current current international technical standard for modelling and is is now not onlytechnical to cope standard with single information exchange between various software servicing cutinformation exchange between various software servicing cutproducts, a limited product range or existing product families, ting tools and holders, as manufacturers tingalso tools andable holders, as well well astobetween between tool manufacturers but to be to analyze andas comparetool products to define and suppliers. It is developed to facilitate the use, manipulation and suppliers. It is developed facilitate that the use, manipulation new product families. It can betoobserved classical existing and of cutting tool within among and exchange exchange ofare cutting tool data data within and and amongormanufacmanufacproduct families regrouped in function of clients features. turing, and turing, distribution, distribution, and usage. usage. However, assembly oriented product families are hardly to find. When the operation of tools based aa standardized forWhen operation of level, tools is isproducts based on ondiffer standardized forOn the the product family mainly in two mat, the information conversion quality and speed will be sigmat, the information (i) conversion quality and speed will be sigmain characteristics: the number of components and (ii) the nificantly improved, and the run In nificantly improved,(e.g. and mechanical, the system system will will run smoothly. smoothly. In the the type of components electrical, electronical). long run ISO boosts exchange efficiency long run the the methodologies ISO 13399 13399 standard standard boosts the the exchange Classical considering mainly singleefficiency products and the quality. At time, it andsolitary, the communication communication quality.product At the the same same time, analyze it essentially essentially or already existing families the liberates the user from any additional work needed to liberatesstructure the useron from any additional work needed to initiate initiate product a physical level (components level) which the exchange various parts of assembly the data data difficulties exchange between between various of the the definition assembly [1]. [1]. causes regarding an parts efficient and ISO 13399 defines the information structure needed deISO 13399 defines the information structure needed to tothis decomparison of different product families. Addressing scribe scribe various various data data about about cutting cutting tool tool assemblies. assemblies. The The standard standard
cc© 2018 2212-8271 The Authors. Published by Elsevier B.V. 2212-8271© 2018The The Authors. Published by Elsevier 2212-8271 2018 The Authors. Published byElsevier Elsevier B.V. B.V. 2212-8271 2017 Authors. Published by B.V. Peer-review under responsibility of the scientific committee of 51st on Systems. Peer-review under responsibility of scientific the scientific committee theCIRP 51stConference CIRP Conference Conference on Manufacturing Peer-review under responsibility of the the scientific of the the of 51st CIRP Conference on Manufacturing Manufacturing Systems. Systems. Peer-review under responsibility of committee 28th CIRP Design 2018. 10.1016/j.procir.2018.03.178
216 2
Darya Botkina et al. / Procedia CIRP 72 (2018) 215–218 D. Botkina et al / Procedia CIRP 00 (2018) 000–000
helps to offer a safe and convenient communication environment for [4]:
about the new assembled tool can be send by using internet of things modularity [1].
• Tool item: version, definition, properties, mutual relationship between each other etc. • Tool assembly: definition, properties, assembled instructions, additional data for 3D model software (CAD, CAM/CNC). • References to external documents: 3D model of a single or tool assembly. • Multifunctionality: tools being used for more than one purpose. • Nominal and physical tools: nominal tool information to CAM and the tool room. Besides, physical tool information between the tool room and CNC. The data about the tool and the tool assembly are broken into the following four major classes, or modules, and Fig. 1 represents an assembly structure of cutting tools [2]. • Cutting item: information about the cutting edge of a tool, i.e. inserts and heads of solid tools. • Tool item: information about tool holders of cutting items. i.e. solid bar or solid head which is used for holding inserts in place. • Adaptive item: information about adapters, i.e. adaptive connections between a cutter and a machine. • Assembly item: information about small joints, i.e. components that join different parts of the assembly together (screws, clamps, etc).
Fig. 2. Concept evaluation, CAM simulation and ToolMaker application environment
In this way, cutting tool data are structured and standardized, which facilitates the connection between all parties of production, and creates 3D models for further applications (see Fig. 2). 3. Information flow 3.1. LISA – line information system architecture
Fig. 1. Different type of items and assembly [2]
This data structure largely resolves the problem of tool classification, for instance, two different tools can be made up from the same “catalog items”, but set to different nominal dimensions, thus there is no danger of confusion at the simulation stage. The verification and implementation of the standard ISO 13399 is provided by a software called ToolMaker. It was developed by Sandvik Coromant initially as an open source application for the practical use of the standard, and currently is updated by KTH Royal Institute of Technology, adding new modules and functions. ToolMaker provides a convenient and efficient interface for the representation of tool information in a structured way (Fig. 2). Users can create a library of necessary items and tools; these files can be used to build an assembly. The assembly can be exported as 3D model and property file. In addition, information
Abundance of communication protocols and formats make information exchange between different levels and stages of production more difficult. For example, many companies prefer to use the point-to-point communication format which necessitates an update of the entire system to include a new application [8]. KTH Royal Institute of Technology initiated the development of a flexible information architecture, called LISA (line information system architecture), that would facilitate communication among equipment of different origin, capabilities or ages. LISA is an event-based service oriented architecture (SOA) [8]. LISA allows individual applications to render themselves as services in the architecture so that other application would be able to interface with them. LISA results in an architecture which keeps applications loosely attached and retains a flexible message structure for an efficient integration of applications. The central elements of LISA are the message formats, communication and service endpoints and the message bus. The message bus, referred to as the enterprise service bus (ESB), enables LISA to transform events into usable data in a sufficiently
Darya Botkina et al. / Procedia CIRP 72 (2018) 215–218 D. Botkina et al / Procedia CIRP 00 (2018) 000–000
217 3
Fig. 3. Demonstrator: Digital twin of a cutting tool
flexible way. An event-based architecture like LISA can manage structural changes in production, simultaneously allowing for diverse applications [8]. 3.2. Tweeting machine Communication between the production line and other possible consumers of information (e.g. process owner, CAM engineers, etc) is implemented in the form of detailed and specific messages called tweets via the so called tweeting machine. The tweeting machine allows to get real-time data from CNC machining by sending tweets. Tweets are published to the enterprise service bus (ESB) by using LISA architecture. These data, transferred to LISA, provide users with the possibility to subscribe to the information. It offers useful and structured real-time data from CNC machining, which can be utilized e.g. to aid in analyzing and decreasing processing flaws, increasing quality, decision making. The Tweeting machine project was coordinated by KTH Royal Institute of Technology and aimed to demonstrate internet of things (IoT) functionality and realize traceability in two manufacturing scenarios: 1) Manufacturing (CNC machining) tracing based on STEPNC. For example, data about the tool being used, its ongoing machining operation, toolpath, forces and torque, time of usage, etc. 2) Quality (product characteristic measurement) tracing based on STEP-NC. For example, axis-positions and data from sensors, using design criteria and tolerances. Thus, the tweeting machine is a powerful tool to get updated data at every moment of production, which can be triggered by the user or other applications continuously. Other applications have the possibility to handle these messages and process the data, update the models and analyze the data. 4. Use case Manufacturing of products for heavy vehicles, for instance, a cylinder head, usually includes the participation of a few par-
ties, as manufacturer needs adapted tool for the production. The companies have to have an efficient method for exchange of information in a way that promotes design and production. The scenario underlying the digital pilot is described below and schematically represented in Fig. 3. 1) An advanced product is designed using CAD, and proceeded to manufacture it using CAM. Suppose that at either of these stages, CAM or CAD, it turns out that a new tool from the tool provider is required for a particular operation. 2) The digital model of a tool is realized based on ISO 13399 and ToolMaker application, using the tweeting machine functionality. The initial values for the process are determined. Extra conditions, for example, available power, forces and torque, machine characteristics and any desired tool change are considered. A reference to the digital “as designed” or “nominal tool” model is tweeted to the database. 3) With product design from CAD, process description from CAM and tool description “as designed” or “nominal tool” from ToolMaker or other tool configuration utility, the machining solution can be compiled, simulated and verified. 4) The tool products are ordered and delivered. The physical tool assembly is built based on the description presented in step 2). The actual dimensions of the finished assembly are measured and tweeted, so that an instance “as realized” or “physical tool” of the digital model can be created. A reference to the digital “as realized” or “physical tool” model is tweeted to the database. 5) At the stage of part manufacturing, all data relevant to the “as used” tool are tweeted. These include measurements of parts, tool compensations and changes, cutting data, etc all with STEP-NC contexts. Here, the previously developed CNC/PLC interface publisher, STEP viewer and receiver applications are utilized for further functionality. An XML-based interface to Productivity Analyzer (Sandvik Coromant’s internal data collection system for customers), is being developed. 6) Analysis of collected data might be implemented with ma-
218 4
Darya Botkina et al. / Procedia CIRP 72 (2018) 215–218 D. Botkina et al / Procedia CIRP 00 (2018) 000–000
chine learning technology and data analytics. 7) All relevant information about tool utilization is visualized for operators. 8) Suppose that a bottle neck operation has been identified and it has been determined that a 15% reduction of its process time is needed to reaching the productivity goal of a manufacturing system. This may be due to an unpredictable tool life, or, for example, because the production personnel decides to to adjust the cutting data to keep measurements within tolerances. In this eventuality, one or more tool vendors are contacted by tweeting a request to help to increase productivity of this operation by 15%. 9) Suppose that a tool provider returns by tweeting a proposal for an optimized process or tool. At this stage, software tools can be used by the company’s specialists. 10) The optimized process is verified and introduced into production. The proposed scheme allows to realize the data flow at every step of production, and ensures the possibility for continuous improvement of process and tool. The following analysis of the data enables a better prediction of behavior, simulations and performance calculation. 5. Results and conclusion The increasing need for an effective data exchange and communication is one of the challenges in manufacturing today. The need of the industry to communicate, share and exchange information is the driving force behind the rapid digitalization of the production line. Engineers require a rapid and wellstructured access to the data collected throughout the whole production lifecycle. A digital twin is a modern solution to these ever-growing needs of digitalization. A digital twin collects data and monitors the process, has access to past data, and, overall, allows for a better understanding of the production process and better prediction of the behavior and results. In this paper, the digital twin of a cutting tool is described. It is based on the international standard for cutting tool data representation and exchange ISO 13399 and the line information system architecture LISA. An information model which collects standardized data automatically, and allows for a continuous adjustment of the digital model has been developed. The adjustment is done by means of a tweeting machine which feeds the data from the production line and helps the evolution of the digital tool from the instance “as designed”, through the instance “as realized”, eventually, to “as used”. The adjusted model accurately represents properties of the cutting tool, and is applicable to precise process simulation, control and analysis, eventually leading to the continuous improvement of the production process. We would like to emphasize two novel paradigms of the proposed solution. First, data collection and flow is based on an international standard ISO 13399 and is maintained during the whole production cycle and tool life. Second, it uses open source applications such as LISA, tweeting machine, ToolMaker, thus opening possibilities for further development. At the same time implementation of the proposed solution poses some challenges which, nonetheless are can not be con-
sidered prohibitive. For example, modern product lines do not typically have the tweeting machine capability: connectivity through tweeting requires modification of existing controllers and PLC-software. 6. Acknowledgements The presented research work is financed by VINNOVA and Powertrain Manufacturing for Heavy Vehicles Application Lab - a Collaboration between KTH, Fraunhofer and RISE. References [1] D. Botkina, M. Xu ISO ToolMaker application development, thesis, KTH, 2016. c ISO 2006, [2] Cutting tool data representation and exchange, report, [s.l.]: 2006. [3] Y. Li, M. Hedlind, T. Kjellberg, G. Sivard, Cutting Tool Data Representation and Implementation Based on STEP AP242, Smart Product Engineering, Lecture Notes in Production Engineering. Springer, Berlin, Heidelberg, 2013. [4] B. Olsson, Communicating Cutting Tool Data Using ISO13399, report, International STEP-NC Demonstration, Bath, UK 2009. [5] R. Rosen, G. Wichert, G. Lo, K. Bettenhausen, About The Importance of Autonomy and Digital Twins for the Future of Manufacturing, IFACPapersOnLine, 48:3, 2015, pp 567-572. [6] B. Schleich, N. Anwer, L. Mathieu, S. Wartzacka, Shaping the digital twin for design and production engineering, CIRP Annals, 66:1, 2017, pp 141144. [7] F. Tao, J. Cheng, Q. Qi et al., Digital twin-driven product design, manufacturing and service with big data, Int J Adv Manuf Technol (2017), https://doi.org/10.1007/s00170-017-0233-1. [8] A. Theorin et al., An Event-Driven Manufacturing Information System Architecture for Industry 4.0, International Journal of Production Research, 2016.