Synopsis of the MBSE, Lean and Smart Manufacturing in the product and process design for an assessment of the strategy “Industry 4.0” Eugenio Brusa Dept. Mechanical and Aerospace Engineering Politecnico di Torino Torino, Italy eugenio.brusa@polito.it Copyright © held by the author Abstract—The industrial product development is currently supplier, as some implementation, like the Word Class managed by resorting to the Model Based Systems Engineering Manufacturing (WCM), already defines and supports [9]. A (MBSE), aimed to decompose the systems complexity, to the comprehensive discussion about the mutual coupling Lean Manufacturing, allowing to achieve the targets of between Systems Engineering, even in its implementation as Quality, Cost and Delivery (QCD), and to the enabling Model Based (MBSE), Lean (LM) and Smart Manufacturing technologies of the Smart Manufacturing. Those three (SM) is herein proposed, by analysing methods, processes, approaches are still assumed completely uncoupled, against the tools applied by each approach. As a result, they look like the evidence of the disruptive power of their mutual and full edges of an ideal triangle, which defines the perfection of integration, as is herein discussed. This integration looks the their full integration for a unified approach to design, to goal to be achieved for a definitive assessment of the so-called strategic initiative “Industry 4.0”, as is currently promoted produce and to deliver. worldwide to improve the industrial productivity. II. CHARACTERIZING THE MBSE, LM AND SM Keywords—Industry 4.0, Model Based Systems Engineering, Lean Manufacturing, Smart Manufacturing, Product lifecycle A. The MBSE and SE development, System Design. To synthetize herein briefly, the MBSE primarily looks at the product as a complex system and helps the designer and I. INTRODUCTION the manufacturer to manage the whole Product Lifecycle The most recent transformation of the worldwide Development. The MBSE allows decomposing the system industrial organization aims to improve the system quality, to complexity, and assuring a complete traceability of the reduce cost, and to finalize the product delivery to the system requirements to functions, of functions to subsystems customer needs [1]. A review of the product and process and components, of subsystems to the built parts, classified design activity, respectively, is currently promoted. To by a part number. This action is effectively performed, by achieve those targets, a straight application of the Systems resorting to some pillars, like the method, the process, the Engineering (SE) to the product development [2], of the tools and the data management [2]. Gemba Kaizen to the process management [3], and of the The methodology includes a preliminary selection of a enabling technologies promoted by the strategic initiative suitable model of the Product Life Cycle, as the well-known “Industry 4.0” to the industry digitalization [4], automation “V–diagram” depicted in Fig.1, and even other ones [10]. [5] and “autonomation” [6], is proposed. The last two approaches are even known as “lean” (LM) [7] and “smart” (SM) [8] manufacturing, respectively. Many companies currently resort to those approaches, although a complete awareness of their powerfulness seems not yet achieved. Particularly, those approaches are wrongly assumed to be completely uncoupled. The SE is often associated only to the product development, although it is intrinsically linked to the process management. The LM is often perceived as a rationalization of the material processing, by neglecting its connection to the product development. Finally, the disruptive technologies supported by the SM are just considered as a progress of tools, more than a mean to Fig. 1. The ‘V–diagram’ used as a model for the Product Life Cycle in the implement the LM and, very seldom, they are considered as MBSE. a relevant part of the SE implementation. Despite that wrong perception, those three innovation levers are tightly This model clearly states the relevant role of the cooperating to face the product complexity, by assuring customer in defining the system requirements and the quality, cost reduction, effective delivery as well as the importance of the stakeholders. The design activity product reliability, availability, maintainability and safety (Application Lifecycle Management, ALM) is somehow (RAMS). Moreover, they allow a suitable interaction mirrored, by level, with the corresponding actions of between customer, designer, manufacturer, maintainer and manufacturing (Product Lifecycle Management, PLM), and XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE links the system conception to its production, through the homologation, when is foreseen, or to product liability and “V” look of the diagram. A key issue of this method is that it RAMS. applies some reusable and digital models. They include a qualitative description of the system behaviour, architecture It is worth noticing that nowadays the MBSE approach and operation (functional modelling) and a quantitative one includes a combined functional and non-functional or (physical or better numerical modelling), based on a dysfunctional analysis to anticipate the prediction of system numerical and mathematical structure. The numerical reliability, since the preliminary design activity [11]. This modelling is exploited to describe the system geometry, to action is made easy by a straight correspondence between the predict its performance, to make a trade-off of its main steps of the product development and those required by configurations, typically by resorting to an heterogeneous the RAMS analysis, as is described in Fig.3. simulation, in which the functional and the physical models CONTENTS, TOOLS AND PRODUCTS are both included. The verification of requirements and the Customer needs and Mission, Scenarios business modeled Contexts product validation even resort to those models to check the COMMON ACTIVITIES ANALYSES TRADE-OFF: Requirement diagram correspondence between product and model, and between Define alternative Requirements REQUIREMENTS solutions and Activity, Sequence, State, Use case ANALYSIS product and customer needs, respectively. optimise Functions and/or Dysfunctions diagrams (Behaviour); Block, Internal NUMERICAL MODELING block, Package (Architecture) FUNCTIONAL The process brings the user to perform the requirement INTEROPERATED MODELS Functional architecture FUNCTIONAL BREAKDOWN STRUCTURE (FBS) ANALYSIS analysis, then the operational, functional, logical and HETEROGENEOUS SIMULATIONS Logical architecture LOGICAL BREAKDOWN STRUCTURE (LBS) PHYSICAL physical analyses, in sequence, to reach a design synthesis. Product architecture PRODUCT BREAKDOWN STRUCTURE ANALYSIS V&V: (PBS) The tools exploited include some typical diagrams, defined VIRTUAL AND REAL Product integration TESTING within a standard language, as the SysML, but even some and design synthesis architecture frameworks, as they are defined, for instance, by HOMOLOGATION RAMS TARGETS several Departments of Defence (DODAF, MODAF, NAF) / LIABILITY or some Space Agency (ESAAF). Particularly, some typical system capabilities, which are exploited in operation, are Fig. 2. A synopsis of the main features of the MBSE approach applied to identified within the architecture framework, through several product development. views of the system, and this helps the designer to define the Functions best solution among those proposed. Functional Functional Hazard Analysis Failure Finally, several tool software are interoperated through a Logical Conditions Analysis platform, which defines a tool chain, including several data Architecture bases, which need an effective data management to share the Logical Reliability Analysis Allocation information, through a careful control of changes introduced Reliability Target by the operators, classified by a hierarchic level. It is worth Physical Architecture noticing that nowadays aside a functional analysis a Physical Reliability dysfunctional is already accomplished in the preliminary Analysis Reliability Prediction technology trade–off [11]. This includes a preliminary Prediction investigation about the system behaviour in presence of Fig. 3. Comparison between activities and results of the functional and classified failure modes in its architecture, thus allowing a dysfuctional analyses. prediction of the system effectiveness and reliability, before that a final configuration could be defined. The analogy between functional and dysfunctional The MBSE offers some typical features to help the behaviors is defined. As the functional analysis focuses on product developer in reaching the goals above mentioned. As the functions, the functional hazard analysis identifies the Fig.2 shows, the two common activities of the trade-off system failures. Similarly, a logical component performs a analysis and of the requirements verification and system logical operation, while in the other analysis it is required to validation (V&V) are deployed by resorting to the three assure a target of reliability, which becomes a real typical analyses of requirements, functions (and operations) reliability performance in the final product, as a commercial or dysfunctions, and physics of the system. More recently, component is identified to physically provide that logical the application to the industrial product and no longer only to operation. the software, suggested of decomposing the functional When the MBSE approach is implemented, a digital analysis into a preliminary identification of functions and model of the whole product is preliminarily synthesized and operations and then of the logical activities performed by the used to predict the product performance in operation. system architecture, thus adding the logical analysis as an Particularly, the FBS, as is depicted in Fig.4, representing intermediate step of the design activity [2]. The language (as the example of a flywheel on magnetic suspension, is used the SysML) provides some diagrams, made standard to be to generate an IBD, for instance, which allows the trade-off shared between customer, manufacturer and supplier. Three analysis [12]. The latter is sometimes converted into a LBS, main graphical products as the functional, logical and or directly into a numerical model, having the same layout, product breakdown structures are created. They allow distinguishing the functions of system, from the logical but including, in addition and within the blocks, some components, describing their operation, but never the mathematical equations, describing quantitatively the commercial products associated, from the product system performance. Numerical simulation is used to define components, which are then selected, among those actually the label data of the commercial components most suitable available on the market. The design synthesis brings to a to be selected for composing the PBS. definition of the whole product integration, tailored to The software tools used to build up the digital model need to be interoperated, i.e. connections must allow a straight transition of information between the tools [13]. Kaizen [3]. It promotes a continuous improvement (kaizen) This is sometimes a bottleneck for the development of this of the process and of the frame within which is actually approach although several solutions are currently available. performed (gemba), through some small and effective They are based either on a tool chain provided by a unique changes, overcoming specific problems or inefficiencies vendor, who assures the products interoperability by design, (muda), identified step by step, by the people involved in the or on some connectors, compliant with some standards like production activity. This leads to a simplification of the the OSLC [14]. process itself, to improve the customer satisfaction, and to rationalize the whole production line (lean production). The Functional Breakdown Structure (FBS) five principles of the Lean Thinking and Manufacturing [15], are applied, since, the main issues of this approach are the value, the value flow, the process flow, the pull production and the perfection of results. Particularly, a specific goal in the material transformation process is making the theoretical time to produce a given element (averaged on the production baseline), known as the “takt time”, as much as possible close to the real time to produce it, or the “cycle time”, to increase productivity and effectiveness [3,15]. The three pillars of the LM are the so–called house- keeping (HK), the identification and elimination of inefficiencies or muda (ME), and the assessment of suitable standards to be repeatedly applied, by the operator, to the Internal Block Diagram (IBD) process (STD). As for the SE, a method can be identified in the practice of Gemba Kaizen. The process management is meant to perform simultaneously two actions, as the maintenance of the existing practices and their continuous improvement. The first rule applied is “Plan–Do–Check–Act” (PDCA), then a coherent standardization follows, and applies the rule Standardize–Do–Check–Act (SDCA). The goals driving those activities concern the priority of quality over all; the Numerical model for dynamic simulation use of data, collected and retrieved by the process, to evaluate its effectiveness, but even to create a base for a statistical analysis; the target of customer needs and satisfaction as a unique and real target of the whole process. Several tools are exploited. A policy is first stated, to define the object of improvement (policy deployment), then people are involved through the Quality Circles, being groups of operators asked to express their useful suggestions about any process inefficiency (QC). Particularly, they must Product Breakdown Structure (PBS) monitor the effectiveness of operations, to reduce the fatigue of operators, by increasing the ergonomics, safety, productivity, quality, and security, and decreasing the production time and cost. The operators express their suggestions, through different means, but all concern the quality improvement, the cost reduction and the delivery enhancement (QCD). Upon the suggestions received, the management defines some standards, and then the operators, who drive their continuous refinement, test them and allow a definitive assessment. When the Gemba Kaizen is applied, several paths are followed, constituting a sort of checklist of activities. They are organized like into a matrix form. The rows of that ideal Fig. 4. Example of the evolution of the MBSE digital model of a flywheel on magnetic suspesion. matrix are the three activities of HK, ME, and STD previously described. They define the items of the process management, somehow like the use cases of the SE. The B. The Gemba Kaizen and the Lean Manufacturing matrix columns are the three main goals defined by the QCD Many approaches currently applied to the process system. They define also the metrics to be applied, to management, more than to the product development, as the evaluate the effectiveness of the running process. SE does, including the Total Quality Control (TQC), or Particularly, when the manufacturer plains the activity, he Management (TQM), the Just In Time (JIT), the Total defines the Quality Function Deployment (QFD, related to Predictive Maintenance (TPM), the WCM already cited, ISO 9000 series and 14000 and others), the Cost metrics basically resort to the Japanese philosophy of the Gemba (about product quality, productivity, stocks, production line flexibility, machinery stops, use of space, lead-time), and The performance of process is easily evaluated, by Delivery targets (efficiency, promptness, completeness, time, filling, along the production line, the so–called Value Stream related to the implementation of the JIT). Map (VSM), in several data boxes, where all the indexes describing the effectiveness of the running process are The maintenance is performed by implementing the certified. housekeeping, and five activities are performed. They compose the so–called set of “5 S” (seizi = clean out the production line; seiton = configure properly what you kept in C. The Industry of the Future and the Smart line; seiso = clean the machinery and check; seiketsu = Manufacturing applied the three above steps to the operators; shitsuke = Proposing in few sentences a complete description of the assure the self-discipline of the operators, write the standards strategic initiative “Industry 4.0”, resorting to the Smart and make some practices). According to that scheme, the Manufacturing aimed to enhance the industrial productivity, rules of housekeeping are defined, and the related standards is rather difficult. Nevertheless, it is known that the Fourth are written. Industrial revolution [4], coming after the introduction of machines, production lines, robotics and automation in the The standardization is even deployed by considering the factories, is based on the smart cyber-physical systems and targets of quality, by resorting to a list of five issues, known the Big Data technologies, which deeply exploit the internet as the “5 M” items (men, machinery, materials, methods, (now Internet of Things, IoT), the cloud, and remote sensing metrics). and monitoring systems. Those enabling technologies are The improvement is based on the elimination of bringing the Industry to the future. inefficiencies or muda, and is performed by identifying the They support the creation of suitable infra- and intra- root cause by answering to a sequence of the so–called five structures to implement the SE and the LM. Smart and “why?” or “5 W”. A classification of muda into mura intelligent systems are widely interconnected, to perform a (changes, variations, irregularities) and muri (excesses), true collaborative and somehow autonomous work, to be respectively, helps in sorting the problems to be solved. They adaptable to the working environment changes, to allow a consider seven typical categories (7 muda), as the excess of continuous and effective monitoring, prognosis, diagnosis production, the excess of stocks, inefficiencies related to and control of systems in operation. product defects, operator motion, process performance, late incoming of goods in production, and transportation systems. To investigate the interaction between SM, MBSE and LM, a short synthesis of the enabling technologies The architecture of the Gemba is even well defined. The characterizing the fourth revolution is proposed in Fig.6, Gemba House, like in a framework, describes it completely according to [16]. [3]. The production line layout is configured upon the principles of the Total Productive Maintenance (TPM) and 1 Advanced Manufacturing Solutions (Collaborative Robotics) the Total Flow Management (retrieving the information back 2 Additive Manufacturing from the customer, as an input to retail units, distribution, 3 Augmented reality manufacturing, and supplier), respectively. Very often, a 4 Simulation (performance, process, machine) structure organized by cells is proposed, to define different steps of the manufacturing activity [15] (Fig.5). MBSE and SE 5 Horizontal and Vertical Integration (Units, Sections, Departments) MATERIAL MATERIAL CELL A DISTRIBUTION CELL B DISTRIBUTION 6 Industrial internet 7 Cloud PROCESS 1 PROCESS 1 PROCESS 1 PROCESS 1 8 Cyber-security 9 Big Data and analytics SEPARATOR SEPARATOR SEPARATOR SEPARATOR Fig. 6. Selection of enabling technologies introduced and enhanced by the PROCESS 2 PROCESS 2 Industry of the Future [16]. SEPARATOR DELIVERY SEPARATOR DELIVERY One of the main goals of those technologies is allowing a WASTE WASTE cyclic use of products, i.e. monitoring and maintenance of the manufactured systems should increase the possibility of re-use or longer use. A crucial issue is the integration of Fig. 5. Example of generic structure by cells of the industrial process as manufacturing units spread on the different locations proposed by the Lean Thinking [15]. (horizontal), with customers and suppliers, as well as that between the design, the management and the workshop, The Gemba includes also a hierarchy of managers and inside the same factory (vertical). operators, all playing a specific and delimited role (to be interpreted as cells of people). The model of Learning All the enabling technologies introduced support an Enterprise, where everybody sees, observes and suggests, is effective enhancement of the manufacturing performance, implemented, through an operational chain starting from the quality and safety, because they are based on the extensive CEO (Chief Executive Officer) and going to the workshop use of both the mechatronics and the digitalized information. operator, through the chiefs of unit, department, and section. The system smartness is often related to different levels of Therefore, the LM exploits a real Training Within Industry artificial intelligence, corresponding to some functions of (TWI) [15]. sensing, controlling and actuating, under a defined strategy [17]. The advanced manufacturing solutions basically include the automated systems and the collaborative robotics, expression of mechatronics, and the additive It might be used as a sentry node of a network to warn the manufacturing technologies, fully based on the industrial operators about any abnormal behavior of either the bearing digitalization of product [18]. components or the hosting system. If it is used remotely, it allows applying the IoT technology, to monitor the life of The collaborative robotics helps humans in making components and warn the manufacturer about any need of faster, controlled and more precise the manufacturing action, maintenance. In case of the active magnetic bearing, the improving the performance, decreasing the pain of operators system simultaneously performs the monitoring action and and assuring high levels of quality and safety. The design of the active vibration control. To install the smart bearing it is collaborative robotic devices surely faces some issues related required a deep description of its calibration and properties, to complexity and to the actual needs to be satisfied, as in the which is digitally provided, since its production, through the exoskeletons. The intensive use of automation in ISO Data Matrix method [22]. Therefore, the smart bearing manufacturing and material processes increases the looks simultaneously as a smart device in operation and a complexity related to multi-physics involved in the coupled smart product in terms of the information contained in its phenomena exploited [19]. Moreover, sensors in automated assembly and shared with the manufacturer, in service. systems allow simultaneously the application of control actions, but even to extract a continuous information from The augmented reality is another effective mean to the operated system, which can be monitored, and analyzed implement the smart manufacturing, as in case of the smart for an effective prognosis of failure and damage conditions, helmet for operators involved in steelmaking or similar as well as for a diagnosis, after that failures occurred. This industrial plants. Basically, this tool provides two services. monitoring action can be connected by the industrial internet The information coming from some sensors embedded and and shared with the operators interfaced with the operating from the network are plotted through a head-up display, and system, or even remotely analyzed, by working units, even read in real time by the user. These data might prevent the far from the location of the monitored system. This use exposure of the worker to some risk or any severe operating involves the transmission of data, through the internet (IoT), condition. Some recent evolutions of this device include a the cloud and under a severe requirement of cyber security. smart glass, allowing to look at the working environment through a glass shield, whose transparency and color can be The additive manufacturing introduces another kind of regulated by resorting to either thermochromic or smartness, related to the digital content of information electrochromic material [23], which might be automatically directly sent by the designer to the production line, activated by a light sensor to protect the user against the risk extensively adaptable to many needs of shaping and of blinding glare [24]. When the operator is required to optimizing the product. It allows manufacturing systems and perform a quality assurance activity in production line, by components previously never built up, because of some monitoring the product, the same device is equipped with surface inaccessible to the tooling machines. The strength of some augmented vision system for damage detection [25], additive manufacturing is the lying of production data which supports the vision activity. It allows detecting directly within the digital product mock-up, made through failures, damages and marks as in gears, rolling elements of the SE as a result of the trade-off accomplished between bearing, or on the surface of the steel strip. technologies. Two examples might simplify the above mentioned concepts. The so-called smart bearing, for instance, is embedded into the machinery as a component of the whole assembly, but is even equipped with some miniaturized sensors, which allow monitoring the inner environment of bearing, to prevent failures and damage, but even the outer and surrounding environment of the hosting frame, as it measures the loading, thermal, vibration and acoustic conditions [20,21]. Sensors Wireless Roller bearings connection Rolling mill vibration monitoring Data Fig. 8. Concept of smart helmet with protection shield based on the acquisition electrochromic smart materials [24] and augmented vision system for damage detection. Data elaboration All those systems exploit a variety of coupled phenomena Vibration and kinematic energy harvesting via piezoelectric / magnetic coupling and include a number of components that their complexity easily rises up and requires some systematic approach to Diagnosis / Prognosis / Control / Maintenance design the device, as the MBSE, and to perform the detection Fig. 7. Concept of smart bearing for large equipment monitoring with of waste, according to the LM approach. embedded sensors and autonomous energy supply. MBSE Model Based Systems Engineering Gemba Kaizen Lean Manufacturing 1 4 PRODUCT Development TARGET PROCESS management 1 5 5 6 Traceability MAIN FEATURES Visual management 5 7 7 8 Reusability Continuous improvement 3 9 3 4 Decompose complexity Apply the Learning Enterprise approach 5 6 1 5 Improve quality GOALS Improve quality 1 5 4 5 Reduce cost Reduce cost 1 3 3 4 Avoid human mistakes in product design Avoid human mistakes in processing 3 6 4 9 Avoid re-engineering Improve delivery process 9 2 4 7 Digital models NEEDS Flexible production line (Gemba) 1 2 6 5 Interoperable tools Self-disciplinated operators 5 7 6 7 Reliable tool chain Reliable machinery 1 9 8 5 Secure Data Base Data retrieving 9 7 5 8 Customer ACTORS Customer 5 7 5 6 Stakeholders Stakeholders 5 6 5 8 Operators Operators 5 8 METHODOLOGY Product Life Cycle Object Gemba (Process cycle) Model of product lifecycle (V, spiral …) Base Model of process: Total Flow Management 4 6 Application Lifecycle Management (ALM) Process maintenance 5 6 5 6 Product Lifecycle Management (PLM) Process improvement 9 3 Methods vs targets Selection of technologies 4 6 Modeling Reliability, Availability, … Housekeeping Quality 9 3 ...Maintainability, Safety (RAMS) 1 9 Trade-off Sustainability Elimination of Muda Cost 3 9 1 5 Deployment Service Standardization Delivery 5 9 4 7 Heterogeneuous simulation 4 3 Verification 3 9 Validation 5 7 Requirement Analysis (System goals, … Process Identification and charcterization of Gemba 5 ...requirements) Housekeeping: apply "5S" 5 6 4 6 Operational analysis (System context, … Planning of process … 4 9 … mission, scenarios, stakeholders) ...(Plan-Do-Check-Act PDCA) 4 5 Functional and dysfunctional analysis… Standardization: apply "5M" 4 5 ... (Functional Breakdown System) 4 5 Logical analysis … Standardization of process… 5 9 ...(Product Breakdown System) ... (Standardize-Do-Check-Act SDCA) 4 5 Physical analysis (Product Integration) Elimination of inefficiencies: apply 3 5 "5 W" (Muda identification) 4 6 Design synthesis Classification of muda - mura - muri… 3 4 ...and problem solving Tools 4 5 Diagrams Driving lists 5 6 Requirements diagrams 5S (Housekeeping); 5M (Quality and standards);… ...5W (Root cause); 7 Muda Context diagrams VSM - Value Stream Map (of Data) Behaviour diagrams (Use case, States,… ... Sequence, Activity) Structure diagrams (Block Definition,… Diagrams 5 6 ...Internal Block Definition, Package) Parametric diagrams Example: Fish Diagram 4 5 Architecture frameworks (views) Procedural frameworks 4 6 (MODAF, AF-EAF, AFIoT, AAF, DoDAF, … TQC - TQM - JIT - TPM - QFD ...ESAAF, MODAF, NAF, TOGAF, UAF) The Gemba House 4 Language Operators team 5 6 UML - SysML - IML - AML - LML Quality Circles (QC) who express visual suggestions Information Requirements, Functions, Components, Parts Standards Numerical contents Value Stream Map Key Performance Indicators (KPI) Key Performance Indicators (KPI) 1 Product baseline Platform Process driveline 1 6 7 Tool chain of interoperated software Operator hierarchic chain 5 6 7 Data bases Data bases 6 7 INDUSTRY 4.0 - ENABLING TECHNOLOGIES 5 Horizontal / Vertical Integration 1 1 Advanced Manufacturing Solutions and Collaborative Robotics 6 Industrial Internet 2 2 Additive Manufacturing 7 Cloud - IoT 3 3 Augmented Reality 8 Cyber-security 4 4 Simulation and system integration by modelling 9 Big Data and Analytics Fig. 9. The proposed synopsis of the MBSE, Lean and Smart Manufacturing. Interfaces (HMI). Actually, both the drivelines exploit all of III. TOWARDS A UNIFIED APPROACH those elements. Moreover, the attention to stakeholders is high in both the contexts. A. A synoptic interpretation As the method is implemented, it can be realized that If one compares the two approaches of the MBSE and the despite the difference of nomenclature and of the context LM actually realizes that a punctual correspondence exists. (product vs process) a certain dualism is present. The ALM That comparison is tentatively proposed in Fig.9. activity is mirrored in the “V-diagram” by the PLM, as in the Particularly, following some typical references as LM maintenance is alternately performed with improvement. [3,7,10,15], the main contents of the MBSE (left column) are The targets are analogous; since the aim of product compared to those of the LM (right column). Each element development is the RAMS as in the process, the quality must of comparison is described in the middle column. Moreover, be assured. The sustainability pursued in the product after collecting the replies to a preliminary questionnaire of development corresponds to the efficiency in process, and 26 companies, the major influence of the disruptive both require keeping cost low. The output of MBSE is the technologies proposed by the SM were associated to each service as a phase of the delivery, being the target of the LM. item, by selecting the two most commonly identified. The legend of numbers and colours is proposed at the bottom of The different steps of process, in both the contexts, Fig.9. express a dualism. In the product development, the analyses are performed in sequence, and in the manufacturing, actions As is evidenced by Fig.9, the MBSE applies to the are executed in sequence, by resorting to a number of industrial product a methodology that is similarly applied to conventional driving lists (“5 S”, “5 M”, “5 W”, 7 muda), as the process by the LM. An almost perfect dualism is well as in the MBSE, the applied language provides several perceived. In some cases a superposition of contents occurs. suitable diagrams. Even in the LM, some diagrams are For instance, the goals are the same, they focus on quality, plotted and exposed in the production line, to involve the cost, mistake, and inefficiencies. In the LM the role of operators in the continuous improvement, as the Ishikawa humans is very evident and the operators are elements of the diagram or “Fish” Diagram, where the targets of QCD are process, like in the MBSE, although they are less related to the 5M at different levels, and to the environment. expressively evidenced. The actors are even the same, and The smallest arms in this diagram are the so–called key customer plays a crucial role. The data are extremely points for the punctual intervention of change (Fig.10). For important in both the drivelines all those activities, the use of tools to implement a heterogeneous simulation is mandatory, as well as the B. Dualisms and analogies support of an effective cloud and of the internet, to allow a Analysing deeply the synopsis, one can find some complete interoperability. The data sharing and management dualisms and analogies. A first evident dualism involves the is crucial, thus requiring a perfect horizontal and vertical requirements of the product development and the standards integration, and to resort to some software deploying the of process deployment. They are both used as a reference for Manufacturing Execution System (MES). the verification and validation, they come out from an Men Machinery Materials Issues iterative process of assessment and refinement, which motivate resorting to all of tools foreseen in the two contexts. Goals The requirement traceability is a key issue of the SE methodology, as in the LM the Visual Management is, i.e. for a continuous improvement the information, the problems QCD and the corrective actions applied must be clearly accessible by all of the operators. For both the digitalization is a crucial target of innovation, as is promoted by the SM, but even the Key point effective integration among units (horizontal and vertical). Environment Methods Metrics Deployment In both the contexts, decomposing the complexity is a priority, in the MBSE simplifying the system architecture is Fig. 10. The Ishikawa or “Fish” Diagram, used in the Lean Manufacturing. mandatory as well as making lean the process is the goal of the LM. The goals even include a difference like the It is worth noticing that in both the contexts, the reduction of cases of re-engineering in the product design, frameworks play a significant role. The MBSE resorts to the and the improvement of delivery, in the process design. They architecture frameworks to deploy the system, in terms of are both focused on the overall process implemented and capabilities and views, as the LM actually implements they promote a unique execution, to keep the costs as low as several procedural frameworks (the Gemba House or the possible. The implementation of the two methodologies of TQC, JIT, QFD) to manage process, materials and time. the MBSE and of Gemba Kaizen look needing a straight use of augmented reality, simulation and modelling, as well as an By converse, it is relevant that the MBSE totally trusts in efficient communication and sharing of information, through the language used to create the digital models, while the LM the internet. directly organizes the operators, both in hierarchy and in groups, or Quality Circles, to retrieve the information and to The needs express a complementarity of exigencies, i.e. support the improvement. Similarly, if one looks at the the MBSE expressively requires suitable tools for modelling, platform applied, the tool chain is dominant in the MBSE interoperated and reliable, based on secure data; the LM while the LM focuses on the operator chain. points out the need for machinery and operators, reliable and very well interfaced, by some suitable Man to Machine Concerning the information, a superposition between the systems (M2M), and more in general by Human Machine two approaches occurs. The elicitation of traceable requirements, linked to the customer needs, corresponds to satisfaction of needs. This can be assured, thanks to flexible the assessment of the process standards, based on customer and lean production lines, as well as by means of smart needs (where customer might be even the following systems and equipment, easily adaptable. The smartness manufacturing unit), but refined step by step through the often increases the system complexity, thus motivating the concurrent contribution of all the operators or the application of the MBSE to decompose and handle it. stakeholders. The Value Stream Map is somehow overlapped to the quantitative contents of data shared in the product What kind of benefits a final integration of the MBSE, development. LM and SM might provide? To this question, some answers are proposed. The use of Key Performance Indicators (KPI) is definitely recommended by both the SE and the LM A. The integration between MBSE and LM shall refine approaches. They define the metrics used to evaluate the and complete the assessment of the Product lifecycle model product and the process, respectively, and provide a list of assumed by the SE. Particularly, it is well known that a link suitable items about which the analysis can be effectively between the ALM and the PLM or PDM (Product performed. In the LM some KPI are frequently used as the Deployment Management) is established by the SE tools, and Overall Equipment Efficiency (OEE), or the Single Minute is currently exploited to clearly define the requirements Exchange of Die (SMED). related to manufacturing. Nevertheless, the SE very seldom defines in details the activities foreseen by the ascending arm At higher level, it might be noticed that as in the SE the of the “V-diagram”, visible on the right, in Fig.1. A clear Product Lifecycle Management is the highest level of the decomposition of the actions after sale, as the delivery, the organization driving the building up of a tool chain to control service, the maintenance are seldom defined, as in some the changes, in the Gemba Kaizen, the Total Flow specialized contribution as in [26], where the introduction of Management drives the strategy of production. It might be a second path looking itself as a “V” is exploited to add the oriented to a “one piece flow”, with a synchronization based personnel training, the maintenance, the monitoring and the on the “Just in Time”, to perform a “pull production” more decommission, as useful actions to describe completely the than to a “push production”, since it is excited by the delivery. customer demand. B. The Gemba looks like a system and, in principle, no The impact on those analogies of the SM looks large, limitation inhibits to apply some of the tools of the SE to the according to the feedbacks collected. If one looks at the process, once that the production line is identified as the proposed association between the enabling technologies and system to be analysed. Particularly, the diagrams exploited the items identified for both the methodologies (Fig.9), by the SysML to decompose the system complexity might be immediately can realize that a good coverage is assured. freely used to analyse the process. Some specialized diagrams, as the State Machine, can be even simulated to Moreover, the contribution of advanced mechatronics, in check the performance of the system [2]. terms of advanced solutions for manufacturing and robotics and augmented reality is relevant and affects both the C. The integration between LM and SM looks natural, if product development and the process deployment. By one assumes that the SM is conceived to enhance the converse, the Additive Manufacturing, nowadays so productivity. Many enabling technologies are required to strategic, provides a good contribution in some issues, while make faster, more effective and more precise the action of the perception of a huge impact on the overall system looks improvement. Nevertheless, all the technologies supporting lower. the monitoring, prognosis and diagnosis activities will provide a key contribution. Particularly, if the remote control The simulation still represents an important element, currently applied to systems in operation, like motor particularly in the meaning of extended heterogeneous vehicles, trains, aircrafts and spacecrafts, will be even simulation, including functional and numerical modelling. applied to the elements of manufacturing systems, for The horizontal and vertical integration seems more a target instance to the bearings, to retrieve data for an effective than an input for the application of such unified approach, maintenance [20], or to the testing facilities, assuring the although a preliminary organization of the working units and system quality, the benefit will increase significantly. of the operators to be effectively integrated is needed, to apply the disruptive technologies above described. It is known that mechanical components requiring a continuous maintenance, being designed for a finite life and All the issues related to the network, the data collection, somehow consumable, need a clear traceability of their elaboration, transmission and management are crucial, for intrinsic and operational data since the testing performed many activities here mentioned. Particularly, the technology before the delivery. Therefore, a real horizontal integration and the infrastructures related to the industrial internet and to with customer will be complete, when the test, the service the cloud is perceived as a key element of powerfulness of and the maintenance will be suitably monitored and coupled. the whole rationale. The impact of the Big Data and This action resorts to the SM smart systems and data analytics is impressive, although the cybersecurity might be, management ass a key element of the infrastructure to simultaneously, the element either of strength or of weakness actuate the remote testing and operation monitoring. of this system. D. The integration between MBSE and SM is defined in C. Towards the integration two levels. If one looks at some smart systems like robots, As it was demonstrated, a relevant issue of the mechatronic and autonomous systems, the system integration convergence among MBSE, LM and SM is the customization is suitably driven by the MBSE, through all its tools. of product. More and more the customers require a Nevertheless, if the activity of remote monitoring is personalized version of product, or better a complete designed, the MBSE is helpful to define all the system parameters, considering the mission, operation and requirements, related to service. Quite often, it happens that For the product development, the main stream of despite the application of remote monitoring systems innovation concerns the application of the digital twin and connected through the cloud, the designer is poorly aware functional modelling in addition to numerical modelling, for about the real specifications required by the application, a comprehensive virtual engineering, prototyping and testing since a too short investigation about the requirements and the [30]. Nevertheless, the effectiveness of those tools depend on functions to be exploited is preliminarily performed. a complete development of the interoperability protocols, of the IoT infrastructures, of the cloud and related services, E. To clarify the mutual integration of the MBSE, LM, needing to be more and more service oriented [31]. and SM, the example of the smart bearing looks suitable. It is first a product to be developed and equipped with a set of Other technologies are strictly involved, as the ICT, with sensors, then it becomes a node of the monitoring network particular care of the network band and configuration, as the and can perform the in-monitoring of its own defects and 5G. In addition, even the HMI systems could improve the failures, as well as the out-monitoring, i.e. it is a sentry of the impact of the proposed approach. A crucial issue concerns process performance for the machinery, where is embedded. the inclusion into the global deployment environment Moreover, the bearing as a system to be tested needs a test previously described of optimized business models, supply bench for a complete homologation. The results of this chains, logistics to configure a balanced ecosystem in the activity are enclosed into the firm of the bearing, nowadays factory. traced, by the labels applied, easily detected and read in operation, according to the ISO Data Matrix [22]. The REFERENCES contents of the data collected and marked on the bearing as [1] J.P. Womack, D.T. Jones, and D. Roos, The Machine That Changed well as their direct transmission to the central data base for the World: The Story of Lean Production - Toyota's Secret Weapon in monitoring purpose can be done by a smart test bench itself, the Global Car Wars That Is Now Revolutionizing World Industry. when equipped with the needed devices. Free Press, 1990. [2] E. Brusa, A. Calà, and D. Ferretto, Systems Engineering and Its This example clarifies the mutual interaction occurring Application to Industrial Product Development. Cham, Switzerland: between the MBSE, the LM, and the SM. Actually, when the Springer, 2018. product bearing is developed, the required testing and [3] M. Imai, Gemba Kaizen: A commonsense approach to a continuous monitoring activities are designed together the system, improvement strategy. The Kaizen Insiitute, 2012. through the PLM, within the MBSE approach. In service, it [4] K. Schwab, The Fourth Industrial Revolution. Cologny, Switzerland: plays the role of system exploited to support the process, as a World Economic Forum, 2016. mechanical component, but even to monitor its performance, [5] Industrie 4.0 integrative production, Aachener perspektiven. RTW as a node of the IoT, thus contributing to the data retrieving Aachen and Fraunhofer, 2014. useful to implement the LM. As a smart system, it resorts to [6] I. Garbie, Sustainability in Manufacturing Enterprises: Concepts, the disruptive technologies of the SM, including the Analyses and Assessments for Industry 4.0 (Green Energy and Technology). Springer, 2016. mechatronics, the IT, especially in terms of cloud, network, [7] J.P. Womack, D.T. Jones, Lean Thinking: Banish Waste and Create data management and storage. Wealth in Your Corporation, Revised and Updated. Free Press, 2003. [8] H. Kühnle, G. Bitsch, Foundations and Principles of Distributed IV. CONCLUSION Manufacturing: Elements of Manufacturing Networks, Cyber- Physical Production Systems and Smart Automation. Springer, 2015. A full integration among the MBSE, the LM, and the SM [9] R.J. Schonberger, World Class Manufacturing: the Next Decade is the natural path for the final assessment of the “Industry of Building Power, Strength, and Value. Free Press, 2013. the Future” strategy. They certainly help in assessing the [10] D. Walden, G. Roedler, K. Forsberg, D. Hamelin, and T. Shortell, required standards, to assure the security and safety levels in Systems Engineering Handbook of INCOSE. 4th Ed., John Wiley and products and processes, compatible with the desired Sons, 2015. sustainability. [11] A. Tundis, M. Muhlhauser, A. Garro, E. Brusa, and D. Ferretto, “Dependability assessment of a deicing system through the RAMSAS The means to perform that integration are currently method”, Proc. 2017 IEEE ISSE, 3rd Int. Symp. on Systems available, or at least are in rapid development. Despite the Engineering. Wien, Austria: October 11-13, 2017. different origins, all those levers for innovation focus on the [12] E.Brusa, D. Ferretto, “Impact of the Model Based Systems same goals. They are motivated by the need of satisfying the Engineering on the design of a mechatronic flywheel–based energy storage system”, Proc. IEEE 2nd ISSE 2016, Int. Symposium on customer, by assuring quality, keeping cost as low as Systems Engineering, Edinburgh, October 4–5, 2016 pp.171–178. possible, improving service and delivery. To assure a [13] E.Brusa, D. Ferretto, A. Calà, “Integration of heterogeneous complete integration, some actions are required. functional-vs-physical simulation within the industrial system design activity”, IEEE 1st ISSE Int. Symposium on Systems Engineering, A full awareness of people about the powerful Rome, September 29–30, 2015, pp.303-310. contribution of the MBSE-LM-SM system must be reached. [14] I. Vagliano, D. Ferretto, E. Brusa and M. Morisio, “OLSC based tool This activity is currently promoted by the educational integration in the aerospace domain: an industrial case”, Proc.IEEE programmes to the digital factory, worldwide proposed, and COMPSAC 2017, Turin (Italy), 4-8 July, 2017. especially by some dedicated competence centres. [15] P.L. King, Lean for the process industries. Taylor and Francis, 2009. To refine the tools of that synoptic system, the disciplines [16] Piano nazionale Industria 4.0, Investimenti, produttività e innovazione. Min. Svil. Econ., Repubblica Italiana, 2017. of mechatronics [27], smart materials [28], and micro and [17] E. Brusa, A. Calà, “Identifying the Smartness of a Mechatronic Coiler nanotechnologies [29] need to be deepened and enriched, by through the 'Systems Engineering'”, Proc. INCOSE Conf. on Systems some new and original contributions. The machine learning Engineering (CIISE 2014), Rome, Italy, November 24-25, 2014, and the artificial intelligence require to be equally developed pp.116–125. and embedded, in the smart systems. [18] E. Brusa, R. Sesana and E. Ossola “Numerical modeling and testing of mechanical behavior of AM Titanium alloy bracket for aerospace application”, Procedia Structural Integrity, 5, 2017, pp.753-760. [19] E. Brusa, M. Mohammadzadeh Sari, “Modeling of the [25] V.E. Marin, “A novel sensory System for the 3D Surface Profiling of electromechanical coupling between crack propagation and Small Complex Objects”, PhD thesis, University of Toronto, 2016, piezoelectric behavior in active layers of smart device”, Mechanics of pp.72-73. Advanced Materials and Structures, 25(2), 2018. [26] A. Garro, “System dependability analysis: main issues and possible [20] E. Brusa, “Development of a sentry smart bearing as a node for solutions”, Proc. CIISE 2014. Rome, Italy: 24-25 Novembre, 2014. connectivity and monitoring of steelmaking system”, Proc. 2017 [27] E. Brusa (ed.), Mechatronics: Principles, Technologies and IEEE ISSE, 3rd Int. Symp. on Systems Engineering. Wien, Austria: Application. New York, USA: Nova Publishers, 2015. October 11-13, 2017. [28] A. Araujo, C. Mota Soares (ed.), Smart Structures and Materials. [21] E. Brusa, “Design of a kinematic vibration energy harvester for a Springer, 2017. smart bearing with piezoelectric/magnetic coupling”, Mechanics of Advanced Materials and Structures, on line since November 5th 2018, [29] E. Brusa, “Design for reliability of micromechatronic structural https://doi.org./10.1080/15376494.2018.1508795. systems” in B. Ekwall, M. Cronquist (Eds.), MEMS: Technology, Fabrication Processes and Applications; pp.67-126. Nova Science [22] ISO/IEC 16022:2006 – Information technology, Automatic Publishers, 2010. identification and data capture techniques, Data Matrix bar code symbology specification. ISO, 2006. [30] E. Barkmeyer, E.K. Wallace, “Reference Architecture for Smart Manufacturing - Part 1: Functional Models”, NIST Adv. Man. Series [23] C. M. Lampert, “Chromogenic smart materials,” Naval Research 300-1. National Institute of Standards and Technology, U.S. Logistics Quarterly, 7, 2004. Department of Commerce, USA, 2016. [24] A. Azens, E. Avendaño, J. Backholm, L. Berggren, G. Gustavsson, R. [31] M.M. Mabkhot, A.M. Al-Ahmari, B. Salah and H. Alkhalefah, Karmhag, G. Niklasson, A. Roos, and C. Granqvist, “Flexible foils “Requirements of the Smart Factory System: A Survey and with electrochromic coatings: science, technology and applications,” Perspective”, Machines 6 (23), 2018. Materials Science and Engineering: B, 119, 2005.