The Internet of Things for Electronics Manufacturing

P C B M A N U F A C T U R I N G W H I T E P A P E R w w w . m e n t o r . c o m VALOR IoT MANUFACTURING— THE INTERNET OF THINGS FOR ELECTRONICS MANUFACTURING BY MICHAEL FORD AND BJARNE MØLLER Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 2 What does the “Internet of Things” mean for electronics manufacturing? Is it a practical solution for the challenges facing the industry today? For the Internet of Things to be possible in manufacturing, a key bottleneck must be overcome: efficient and standardized communication from machine to machine and from machine to human operators and management. Standardized data exchange over distributed servers to many different access points is what makes the Internet effective and useful. Similarly, for an “Internet of Manufacturing” to work, each production machine and process has to be able to distribute information quickly and efficiently to provide clients with whatever information they may need. A built-in standard communications engine at each SMT-related process is required already because many different messages can come from different machines that represent a single event of significance. All of these messages need to be collected in real-time to be interpreted and create the knowledge of the event. Using a single protocol and format standard for all of the machines would improve control and visibility of operations throughout the manufacturing process, and business intelligence tools would be easier to implement. This is where electronics manufacturing is heading; however, a critical issue is the support of the thousands of different legacy machines that are in use throughout the industry. Without additional revenue, machine vendors are unlikely to modify old machine hardware and software to be viable for the Internet of Manufacturing (IoM). This is where a dedicated external IoM device is needed to implement standards, protocols, and computerization for the whole shop-floor, without having to replace every machine. Mentor Graphics is providing the Valor IoT Manufacturing network platform to enable adaptation for the future while bringing legacy machines along in this evolution. It consists of a robust hardware device with embedded software that supports live bidirectional data flow between all electronics manufacturing shop- floor machines and processes, and it uses the new Open Manufacturing Language (OML) communication standard (www.omlcommunity.com). This network platform is “plug and play” and requires minimal change to the manufacturing environment and work processes. WHAT ARE THE TRENDS PRESSURING ELECTRONICS MANUFACTURING? Trends that have been building gradually for a long time in the electronics industry have now reached a critical mass—putting significant pressure on the need to adapt. Over the past 30 years or so, manufacturing moved to more remote locations away from the market to reduce labor costs. However, with the increase in technology and the increased capability of automation in recent years, many more shop-floor processes can be automated, replacing operators and leveling the balance in costs of manufacturing between local and remote locations. Valor IoT Manufacturing unit. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 3 Another development is the distribution of products from the factory to the customer, whether business-to- business or direct shipping to the customer. Again, this is a trend that has been building over many years because the cost of physical stock and related logistics, with the risk of depreciation, has driven businesses to reduce the number of products present in the distribution chain and the time for the product to reach the customer once shipped from the factory. This shift has been pushed along with direct ordering and Internet shopping. As stock levels in the distribution chain are reduced, factories have to be able to adjust quantities shipped and delivery dates to match demand, which means more flexibility of manufacturing and shorter order-to-delivery lead-times. As more electronic technology is offered that is individualized to each buyer, many variations or configurations of products are expected and demanded by the market. Ironically, being able to provide a greater number of discrete products leads back to increasing stock and the associated costs in the distribution chain. Almost every electronics manufacturing operation today is affected by these market trends, manifesting as greater product mix, demand volatility, and the need to maintain operational performance of automated processes as the number of changes between products increases. Solutions to these pressures such as Industry 4.0 and Smart Factories that have been proposed are orientated around automated processes, such as the SMT placement machines, and other related processes such as inspection and test. These are controlled by computerization, which, in effect, joins each individual automated process into a complete automated factory. Several computerization ideas have been suggested, such as control of production flow optimization, control of materials logistics using Lean pull signals, and closed-loop process adjustments. From a business perspective, these activities are essential, especially when combined with increased automation, to make the manufacturing operation perform with a far higher level of productivity while being more flexible. The combination of flexible automation and higher level control through smart computerization is the key for success in these changing times. Trends leading toward the “Internet of Manufacturing” highlight the lack of standardization in communication within PCB electronics assembly. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 4 COMMUNICATION IS THE BOTTLENECK Mentor Graphics commissioned research to help understand what specific issues have been preventing PCB electronics assembly manufacturers from achieving their goals during these trying times. This survey reached nearly 500 original equipment manufacturers and electronics manufacturing services companies. The questions explored performance objectives, supply-chain issues, capacity concerns, global operational challenges, quality, changing delivery demands, industrial engineering, and process technology, as well as information technology. The results were surprising. All of the traditional issues related to engineering, quality, and business practices trailed behind machine communication challenges. The problems are the access to information from machines and data accuracy, that is, the quality and usefulness of the data from the machines. Whether high- level computerization or simply visibility of the operations, the frustration is not being able to see and understand what is happening. No visibility of issues leads to the inability to make targeted improvements. The implementation of computerization, that is, software control over all automation, as recommended by “Smart Factory” or Industry 4.0 specialists, depends on accurate and timely data from many sources that is available on demand in a live operational environment. It is one thing to have limited access to data where it is a management team discussing, researching, and collaborating to find solutions to problems or to take improvement initiatives. Automated decisions made by computerization can only be based on the available data. A single flaw or omission in the data would be likely to produce a significantly flawed result. CURRENT CHALLENGES TO USEFUL DATA ACQUISITION In the the electronics industry, around 80 to 90% of the materials used to make any product are SMT materials mounted onto a PCB. Many processes make up the complete SMT process, including solder-paste screen printers, glue dispensers, SMT placement machines, automated optical inspection (AOI) of solder paste and component mounting, reflow ovens, and X-ray inspection of reflowed or soldered PCBs, as well as the many varieties of manual operations, such as PCB assembly, system assembly (box-build), in-circuit test, functional test, and personalization, etc. Each of these different process types has a lot of variation. Manual processes have variation in the operational flow and setup of the stations, as well as the exact nature of the work done. In automated processes, many different types of machines use different mounting technologies, made by different vendors. The issue of data acquisition from this vast array of sources is a significant challenge in electronics manufacturing. Until the recent introduction of the Open Manufacturing Language (OML) for PCB assembly (www. omlcommunity.com), no one standard for data acquisition or control has been successful in the electronics industry, and this has hindered progress toward higher levels of automation. Data capture has instead been done using machine native proprietary systems, with little commonality through standardization. Challenges of electronics manufacturing. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 5 All of the different sources then need to be connected to create the complete build record of production—a complex, multi-layered problem, which extremely few shop-floor IT solutions have ever been able to address. The following is a typical example of the steps that have to be taken to gather useful data from a shop-floor process, without implementation of Valor IoT Manufacturing. 1. To obtain data from a machine, a computer needs to be connected to it with an Ethernet cable or other data connection such as RS-232, USB, and electrical triggers. In many cases, the available data from the machine interface may be insufficient, so sensors will need to be added to key points of the machine and supporting mechanical units, such as the conveyor. In some cases, no advanced interface is available, and all information needs to be derived from sensors. Cable specifications, pin configurations, earth connection concerns, and power supplies, etc. are all a part of the complex design necessary to create each of the individual hardware connections to automated shop-floor processes. 2. Once a physical connection has been made, a protocol of data exchange has to be observed. This is usually different between machine types and almost always different between machines from different vendors. The connection required can also vary within machines of the same type that are running different operating software revisions. Establishing the communication, selecting data to collect, and acknowledging messages from the machines all has to be performed exactly and with strict timing. The machine will often not check whether the data has been transmitted successfully because it has to carry on with its placement operation, so a second attempt will not be possible. Issues with protocol may not be apparent even months after installation, and they are often intermittent. 3. Once data is being correctly received from the machine, the data needs to be converted, or normalized, into a standard language. Across all of the various machines, important pieces of information can be represented in many different ways. The data often doesn’t have a one-to-one relationship so many messages from the machine need to be read and evaluated. Then, depending on the content and timing, the messages have to be translated into a single normalized message. The patterns in the data can fluctuate as software revisions of the machine change, which means that issues with data accuracy and reliability frequently occur. 4. Once a single normalized message from a machine is successfully transmitted through a cable, then the protocol, and translated into a standard meaning, it may still not be of value. An example is when a Challenges of shop-floor data collection. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 6 machine in a line stops because a PCB hasn’t arrived at the machine entry point. The message from the machine would be something like “Waiting for PCB”. In reality, reports from machines, irrespective of the interface, mainly consist of these types of messages, which can be as much as 80% of the total data. Machines can only explain their internal state without any reference to cause or external influence. In isolation, this information provides at best the knowledge of absolute asset utilization. Without any associated causes, the responsibility of the event cannot be determined and no indication is given about what action should be taken to resolve the issue and avoid it in the future. Then, in this example, an additional step needs to be taken outside of the machine to examine the data for the preceding process. This is only possible if the data from the preceding process has also been normalized and is available in the same system. The preceding machine may have been stopped because of a “Parts Pickup Error” related to a certain material. As a result, both machines were stopped. The accountability for any stops because of this reason should be inherited by all later machines affected. The preceding machine may have started up again so that the pickup error is now historical data, even though the machine has not yet completed a PCB for the next machine to start work on. The analysis of data between machines is highly time-sensitive. 5. This example may still not be representative of data that is completely useful. The earlier machine in the line reported a material pickup error, but that error can be a result of several different causes. For example, the material supply in the feeder was exhausted the feeder was jammed, or even that the size or shape of the material being picked up was not as expected for some reason, such as a counterfeit material on the reel. The actions to be taken for rectification in each case are different, so qualification of the raw machine message is important. A qualification process may discover that the material may have exhausted because of a material logistics issue, or it may have been removed because of an MSD issue, it may be damaged, or it may be unexpectedly missing. Information from the materials logistic system would help confirm which of these is true. Only with qualification can the message from the machines become a complete item of information. Depending on the purpose of the use of data, different levels of qualification may be needed. However, situations often happen where partial qualification is acceptable initially for a certain application; but then, when another application is introduced later, the same data is seen as useless, thereby requiring significant upgrading of the recording system. The speed and scale at which these messages can be gathered from the machines can be extreme. Each SMT machine is capable of placing several components per second, and each completed electronic product contains potentially many thousands of individual components. This example of data capture is just one of many such examples of the challenges associated with the diverse equipment used in an SMT assembly operation, and why it has been so difficult to address satisfactorily. This complexity has led to the situation that the market research confirmed; one in which information from the electronics manufacturing shop-floor is unreliable, inaccurate, and of limited value. THE SOLUTION: A STANDARDIZED NETWORKED PLATFORM FOR THE SHOP-FLOOR The Valor IoT Manufacturing solution solves these issues of gathering live information from the shop-floor by combining data acquisition and normalization in a single piece of hardware. Within the hardware, embedded software is included for advanced interfaces to SMT machines as well as related equipment such as test and inspection machines. The system also includes a wide selection of hardware interface ports, providing a single connection solution on a “plug-and-play” basis to virtually every machine on the shop-floor. The Valor IoT Manufacturing solution takes care of selecting the correct physical communication parameters and applying the correct protocol. It then automatically acquires the data and converts the various patterns of data coming from the machine into a standard normalized format, represented with OML. This combination creates a platform in which any machine can be connected that will work without modification or specialist setup. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 7 INDUSTRY 4.0 REQUIREMENTS VALOR IoT MANUFACTURING PLUS OML Independence of the communication technology from manufacturer, sector, operating system, programming language. The use of OML is a vendor-independent open standard for vendors and end-users. OML support can be implemented in many languages. Valor IoT manufacturing SDK is supplied in .NET and Java options. Scalability for integrated networking. Horizontal and vertical communication across all layers. Valor IoT Manufacturing and OML can scale to provide any infrastructure needed, data distribution, filtering, and intelligent routing of data to support the deployment of a “big data” collection infrastructure. Secure transfer and authentication at user and application level. Valor IoT Manufacturing uses TLS encryption for transfer of data. SOA, transport via established standards such as TCP/IP for exchanging live and historic data, commands and events (event/callback). OML is independent of the transport method. Currently, two protocol bindings are available: optimized TCP- based binary protocol for high-performance applications and HTTP/HTTPS web service with JSON coded messages. Additionally, producer/subscriber communication model can be integrated. The use of OML guarantees consistent transport of all data. Besides live and real time data, historical data and their mathematical aggregation are also standardized in OML. Mapping of information content with any degree of complexity for modeling of virtual objects to represent the actual products and their production steps. OML provides a fully networked concept for an object- oriented address space (not only hierar¬chical but full- meshed network), including metadata and object description. Object structures can be generated via referencing of the instances among each other and their types and a type model that can be extended through inheritance. Because servers carry their instance and type system, clients can navigate through this network and obtain all the information they need, even for types that were unknown to them before. Unplanned, ad-hoc communication for plug-and- produce function with description of the access data and the offered function (services) for self-organized (also autonomous) participation in “smart” networked orchestration/combination of components. OML defines different “discovery” mechanisms for identification and notification of OML client and their functions within a network. OML producers can be located local (on the same host), in a subnet or global (within enterprise). Aggregation across subnets and intelligent, configuration-less procedure (e.g., Zeroconf) are used to identify and address network participants. HOW THE VALOR IoT MANUFACTURING NETWORK ADDRESSES INDUSTRY 4.0 NEEDS Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 8 The Valor IoT Manufacturing unit is designed to work in the electronic shop-floor environment. Its rugged construction ensures reliability in physically demanding places and susceptibility to dust, heat, or vibration. Each unit has a backup power supply to actively detect power outages, record them correctly, and then initiate a controlled shutdown. The series of machine events and associated data is retained within each Valor IoT Manufacturing unit for as long as three days so that data is not lost in the event of end-user systems or networking issues. The device is robust and resilient to error, and it can be relied on for the most sensitive and demanding applications. COMMUNICATION IS FACILITATED BY A COMMON LANGUAGE The use of Open Manufacturing Language by the Valor IoT Manufacturing unit enables data collected from all of the different types of processes to be represented in a single data format, which can be read and used by many different computerizations or IT systems simultaneously. The use of OML also allows support for extended data collection. Externally developed software can create OML data, which Valor IoT Manufacturing will incorporate, manage, and integrate into the overall shop-floor Valor IoT Manufacturing network. An OML software development kit (SDK) is provided for this purpose. STRUCTURE OF THE VALOR IoT MANUFACTURING SYSTEM As well as providing the data acquisition and normalization function at each machine and shop-floor process, the Valor IoT Manufacturing system provides a distributed processing architecture. A Valor IoT Manufacturing unit can act as a line controller, coordinating the data from many other units attached to machines and processes on the production line. In addition, a unit can be used as a factory floor gateway that manages the connections between the enterprise applications which use OML data and the sources of data from the machine connections. This infrastructure, independent from the generic factory networking system, supports many thousands of simultaneous connections, offering a high degree of scalability. The architecture is designed to cope with many parallel real-time flows of huge data volumes, eliminating potential networking bottlenecks and connectivity issues. This is done by the establishment of point-to-point connections that are managed by the gateway. For example, a test machine may create huge test result files, which could include several high- resolution pictures and a long set of diagnostic results. This large dataset can be automatically routed to a dedicated server that deals with the specialist storage of the information, perhaps as part of a “big data” cloud solution, without affecting the real-time performance of the infrastructure as a whole. Enterprise applications access the Valor IoT Manufacturing information through the use of the supplied SDK. Each of the varied applications, whether a part of an Industry 4.0 or Smart Factory computerization, MES, ERP, a Lean supply-chain solution, or a production flow control computerization, will need only one format and source of information that includes data from the whole of the shop-floor, all of which is coordinated through the gateway. LINE CONNECTIONS The Valor IoT Manufacturing gateway has visibility and control of all instances of the line controllers and the different interface connections that are supporting equipment within the line. The Valor IoT Manufacturing interfaces are capable of supporting a wide range of line equipment in addition to the machines, such as fixed scanners to read unique IDs of PCBs as they arrive or leave a process so that product WIP and enforced routing control can be implemented. Hand-scanners can be connected for material verification or system serial-number scanning, and sensors can be placed on the production line equipment or within the machine, such as in light towers or conveyers, to capture knowledge of events where machines are unable to provide the necessary detail through the proprietary interfaces. Through this flexible infrastructure, the Valor IoT Manufacturing connection can be applied to whatever data acquisition task is required. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 9 In addition to data acquisition, the Valor IoT Manufacturing unit can communicate data back to the machine, which may include commands such as to stop or start or change programs. This communication enables poka-yoke solutions as well as remote-control and management solutions. For machines that do not have the capability of receiving commands, simple start/stop and flow-control mechanisms may be implemented using the hardware control. “PLUG-AND-PLAY” DEPLOYMENT When a new Valor IoT Manufacturing unit is connected to the network, it goes through a simple connection flow. The unit automatically contacts the gateway to receive the simple designated configuration: location on the network, its role which can be data collection or line controller, and information if required about the machine to which it is connected, as well as any peripheral devices such as scanners. The connection is then registered by the gateway, which means that data from the process is available to those who request it. In the ready state, data is collected continuously and stored within the Valor IoT Manufacturing device. An application can request the information immediately as live information, or it can collect a series of past events as part of a batch. MANUFACTURING DASHBOARD Although information collected by the Valor IoT Manufacturing solution is intended for use in several sophisticated computerization systems, a fully configurable web-based dashboard is included with the platform, through which details of individual machines, lines, or the whole shop-floor can be seen. Several key points of interest (KPIs) can be displayed across many different aspects of production operation, including status, downtime, product flow, test results, and statistics. This simple dashboard can support many shop-floor operational and management requirements. The supplied OML SDK allows for any additional scope for reports and dashboards to be developed according to the exact needs of the business. Line connections between the Valor IoT Manufacturing system and machines on the shop-floor. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 10 A SNAPSHOT OF THE VALOR IoT MANUFACTURING TECHNOLOGY IN OPERATION The Valor IoT Manufacturing solution consists of mainly five building blocks. At the top, the Valor IoT Server manages the factory configuration and holds the Manufacturing Process Definitions. The next level is the Factory Gateway, which is the interface between the Valor IoT Manufacturing network and the world around. Below that, the Line Controller gathers and controls data from the individual processes in the line. Then, the Process Manager handles the equipment driver, and finally the I/O Manager takes care of equipment and peripherals connectivity and control. The Valor IoT Server handles the central system and factory configuration and gives the administrator an overview of all Valor IoT Manufacturing points in the factory. It includes a central Valor IoT Manufacturing firmware update function, allowing the administrator to manage the updates on all Valor IoT Manufacturing points in the factory, as well as forcing updates of firmware on the selected Valor IoT Manufacturing points with a forced restart of the point. A central time server ensures that all Valor IoT Manufacturing points are fully in sync. The Valor IoT Manufacturing Server is also the master of the released Manufacturing Process Definition and includes comprehensive API to feed and get data from the server. Backup and maintenance tools ensure backup of vital data from each Valor IoT Manufacturing point and support incremental backup of data on each Valor IoT Manufacturing from the last backup. The main function of the Factory Gateway is to handle the subscriptions coming from external applications and routing all relevant data between Valor IoT Manufacturing points and subscribers. In parallel, the Factory Gateway provides information to the process point coming from the Valor IoT Manufacturing Server, including the factory layout, process point configurations, and manufacturing process definition. It holds the factory- level dashboards, which allows managers and planners quick web access to vital real-time KPIs that show the performance of the factory at any given time. The Line Controller is a combination of hardware and software. The hardware includes a switch with multiple PoE LAN connectors that allows easy connection to the process-point hardware in the line. The Line Controller Dashboard for the Valor IoT Manufacturing network platform. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 11 understands the nature of the line and the included machine and process types and controls the line accordingly. It understands the bottleneck processes. Line-level dashboards provide all information related to the line and allow drill-down into process-level dashboards. The most important component in the solution is the Process Manager. This is where the interface to the process/equipment is handled. With dedicated state machines, full understanding of the process is obtained and converted to the neutral OML language. Equipment interfacing takes care of the machine recipe with actual programs and detailed recipe content, captures real-time status-events (working, loading, waiting, stopped, etc.), handles bidirectional verification events, captures traceability data, and captures process/ quality data such as pass/fail, symptoms, and defects. The Process Manager captures current machine status, measures and saves the cycle-time on each produced board on every machine/module/lane, captures pick-up errors from the machine, enables machine performance displays in real-time, enables feeder (slot) performance display in real-time, and enables number of placed components per hour. A vital part of seamless connection to the shop-floor is the ability to interface with peripherals around the process. The I/O Manager provides full control of PCB ID or unit ID reading in front of the process point. This The building blocks of the Valor IoT Manufacturing system. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 12 WIP point scanning feeds the readings via OML to the subscriber and controls the SMEMA so that the unit is not allowed to enter the process unless that consumer application accepts it. One or two scanners may be connected per lane on the machine. These scanners are triggered by connected PCB sensors. Logic in the I/O Manager sends trigger events to the scanners and keeps the trigger event high until the entire board/panel has passed a sensor. As part of the logic, the sensor input is managed so that small holes in the panel are ignored; for example, a drop in the sensor input for x milliseconds is ignored if the sensor input gets high immediately after. The sensor input management also can keep the scanner trigger high for a number of seconds after the panel has passed the sensor. The ID reading supports 1D, 2D, and RFID, and scanners can be connected via RS232, USB, or LAN (PoE). The I/O Manager connects to handheld barcode scanners via RS232 or USB, which allows easy integration to applications that require material ID, part number, lot number, operator ID, WIP, work order, equipment ID, station, track, or similar readings from the shop-floor. Additionally, external acoustic and visual alarms may be controlled, which gives the application full control of operator interaction and warnings. The I/O Manager also has a fully flexible control of a set of relays and opto-isolated inputs, which gives the application the ability to connect to light towers to understand status of a process, sensors to example monitor simple WIP, and door sensors to do smart verification based on checking if the cover has been opened. The I/O Manager stops the machine in case the connection to the subscribing application is lost, based on a heartbeat check between Valor IoT Manufacturing and the application. Valor IoT Manufacturing can alternatively be configured to let the machine run as long as possible even after missing connection to the application. In this case, all machine events are stored on the Valor IoT Manufacturing unit and can be retransmitted to the application when the connection is re-established. The I/O Manager will only allow the machine to run until verification or an enforced routing event occurs. The actual stop of the machine is done by sending a software command to the machine and/or by interfering with the machines electrical system (normally, the cycle stop circuit), and thereby stop the machine. Although, in many cases, a software stop is preferred because it eases the installation and does not require interference with the machine electrical system, the hardware stop is an important function. It prevents the machine from producing if the Valor IoT Manufacturing unit or the application is down. The Valor IoT Manufacturing device has a “keep alive check” function running, which constantly communicates with the application. In casepower is lost on the Valor IoT Manufacturing unit or the control software does not send keep-alive signals, the machine will be stopped. MACHINE COMMUNICATION The Valor IoT Manufacturing system supports multiple ways to communicate to the processes. On a high level, it uses four main approaches: • Connection to an API on the machine-vendor line computer. • Connection to an API on top of the vendor MES solution. • Direct communication with the single machine. • Flexible machine connection. CONNECTION TO AN API ON A MACHINE-VENDOR LINE/FACTORY PC The Valor IoT Manufacturing unit supports connection to the machine through a standard machine vendor API, where the machine vendor API supplies all machine communication for a full line or even the full factory. Connection to most machines is going through a line-level or factory-level API provided by the machine vendor. Valor IoT Manufacturing connects to this API instead of connecting to the actual machine. CONNECTION TO AN API ON TOP OF THE VENDOR MES SOLUTION Many machine vendors provide their own MES capabilities. This includes functions such as machine performance monitoring, material verification, and material traceability. Valor IoT Manufacturing supports interfacing to these systems. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 13 DIRECT COMMUNICATION WITH THE SINGLE MACHINE Valor IoT Manufacturing supports the scenario where it acts as the host for the machine and sends host messages in parallel with passing-through standard host/machine messages. FLEXIBLE MACHINE A flexible machine interface is available in cases when no advanced machine interface is available for the machine model being installed. The interface can be used on almost any machine (screen printer, THT placement machines, glue dispensers, coating, reflow ovens, and others). It provides machine performance and WIP monitoring and supports a platform for installing component verification and traceability solutions. It requires no direct connection to the machine-control software and relies on hardware to detect machine events. HOW VALOR IoT MANUFACTURING CAN BE APPLIED With the installation of Valor IoT Manufacturing connections on the shop-floor, t several different ways can be used to collect the data collected, including simple reporting and data archiving, as well as more sophisticated computerized solutions based around Industry 4.0 and Smart Factory control. Examples of application that can use the data include: Dashboards Information can be displayed such as production run-rates and achievement versus schedule. Process downtime can be broken down by reason and responsibility. Supply-Chain Information Reports of the consumption of materials by each process, the logistics jobs, and status of materials, for example, arriving at machines or moving back to the warehouse. Spoilage information can also be used, with reference to the cost of materials. Product Tracking Information relating to PCBs that arrive at or leave a process can be recorded to provide visibility. Production WIP and associated issues can also be reported, including the analysis of bottleneck processes, knowledge of the minimum cycle times, analysis of logistics and waiting times, buffer control, first-pass yields, repair loop utilization, and control. Traceability A record of all materials used can be made, including those added by automated processes such as SMT, as well as by manual assembly processes. Parts replaced at repair can be included. In addition to materials, the full history of process data can be stored, for example, process setup parameters, actions and results, including environmental information, and the use of tools. The combined materials and process traceability create a complete product build record. Conformance and Compliance Records of compliance of certain key operations by certain roles on the shop-floor can be enforced and recorded through the Valor IoT Manufacturing solution. This effectively builds in management protocols mandated by the relevant industry sector and the nature of the product to the operation as part of the operation with poka-yoke control. The following are examples of how the Valor IoT Manufacturing platform can be used with Smart Factory/ Industry 4.0 principles. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 14 Finite Planning and Work-Flow Control The live assignment of products to line configurations is a key element of maintaining high levels of productivity in a high-mix environment. With a clear live status of production completions and delivery, a high-level planning and control computerization can decide how to best fulfil production needs according to a dynamic delivery demand. Products can be reassigned between line configurations so as to best meet the delivery needs, with feeders being dynamically grouped between successive products to reduce the changeover times. Finite planning can also use materials information to ensure that all work-orders can be executed and fulfilled without unexpected material shortages. Lean “Just-In-Time” Materials The live consumption and spoilage data can be used to generate a pull signal for the supply and replenishment of materials on the shop-floor on a “just-in-time” (JIT) basis. The support and tracking of material events such as unique identification labeling, put-away, MSD management, logistics, and verification means that inventory accuracy and location control can be assured, removing the need for shop-floor material advance preparation and excessive material buffer stock. Closed-Loop Control Systems Measurement of the performance of production operations can be measured and controlled. Machines in the line can be dynamically reassigned to adjust the balance of placement times when a bottleneck in the process occurs. Another example is the gathering of x and y placement drift data as measured by an AOI machine. Knowing the source of where the placements are made, with the analysis of the pattern of drift, actions can be taken to highlight potential issues before defects occur. Process parameters can be dynamically reassigned to ensure that placement performance remains within control parameters, avoiding defects and downtime. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 15 THE VALUE POTENTIAL OF VALOR IoT MANUFACTURING IT projects in the manufacturing environment need to meet the technical and business goal of providing a compelling return on investment. When Valor IoT Manufacturing is implemented, a return on investment through the reduction of efforts made on the shop-floor to gather useful and reliable data is immediate. However, the values and benefits of Valor IoT Manufacturing can be considered from the many perspectives of different areas within the manufacturing organization that will benefit from the information supplied. The following are examples of the benefits to teams within an organization. FROM THE PERSPECTIVE OF THE PRODUCTION MANAGER/OPERATIONS TEAM The key value for this group is to maximize productivity, with consistent performance and quality. Their benefits from using Valor IoT Manufacturing include: • Early warning of conditions that threaten productivity, such as the indication of significant events and trends, including machine stoppages, changeover times, completion rate changes, and quality events as indicated by first pass yield (FPY) or other KPI, where target is not achieved, or there is a sudden change. • Risk of missing delivery deadline through material or resource issues and significant equipment breakdowns. • Identifying bottleneck lines and operations. • Creating consolidated data for production performance reports. • Guiding of key compliance and conformance operations. • Assuring correct process setup (documentation, programs, materials, tools) verification, poka-yoke control. • Automated recording of traceability data (process, materials and operation). • Missing or late material delivery or resource availability. FROM THE PERSPECTIVE OF THE SUPPLY-CHAIN TEAM The responsibility of the supply-chain team is to ensure the correct and on-time delivery of materials to each of the production processes. Key values include the visibility and guidance of logistics operations, reduced handling and operation workload, and reduced material-related costs. The benefits they can get from Valor IoT Manufacturing include: • Enhanced management of logistics tasks and management of material storage and supply. • Visibility of remaining material and production performance/progress to drive material deliveries (JIT). • Automated assignment of logistics tasks among the logistics teams. • Logistics dashboard that gives visibility and management of overall logistic task schedules, including live view of material logistic task progress, display, and analysis of material delivery metrics, potential process starvation instances, material inventory costs, and logistics traceability. • Accuracy of inventory enables smaller warehouse with greatly reduced excess buffer stock. • Improved and automated input for MRP and ERP. • Accurate allocation of materials by customer/vendor/supply type. • Improved MSD control and reduced handling. FROM THE PERSPECTIVE OF THE QUALITY TEAM Values for the quality team include reduced diagnosis and repair times, with reduced defects through active quality management. The benefits they can get from Valor IoT Manufacturing include: • Enhanced visibility and management of process exceptions. • Direct automated collection of electronic repair tickets. • Enforced routing, including repair loop management. • Enhanced analysis of the repair operation. • Process, materials, and engineering data history enabling rapid issue diagnosis. • Root-cause feedback to production processes and management. • Automated traceability data collection, including process data, materials, and tracking history. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing w w w. m ento r.co m /v al o r 16 • Active quality management based on timely, accurate test and repair data and test results. • Ability to automatically quarantine materials and processes and stop live production. • Advanced quality reports, dashboards, and alerts. • Use of detailed process, materials, and tracking history to identify trends as well as one-off defect causes. • Automated audit process. • Clear and immediate records, including proof of operation, to show conformance and compliance to management standards, including ISO, FDA, etc. FROM THE PERSPECTIVE OF THE PROCESS ENGINEERING TEAM Values include increased effectiveness of equipment programming and creation of other engineering instructions through use of operational feedback. The benefits they can get from Valor IoT Manufacturing include: • Visibility of actual versus expected performance. • Visibility of the minimum and average execution times. • Enhancement of programming and line-balancing parameters to increase productivity. • Enhanced test capability through analysis of test result trends. • Enhancement of DFM rules through analysis of capabilities and related issues. FROM THE PERSPECTIVE OF THE PLANNING TEAM Increased asset use from execution sequence and planning perspective can be gained by using Valor IoT Manufacturing data. The value for this team from Valor IoT Manufacturing includes: • Understanding actual status of production with predicted progress. • Visibility of expected completions versus plan. • Identification of bottlenecks. • Ability to react rapidly in response to changed delivery requirements or issues such as equipment breakdown. FROM THE PERSPECTIVE OF THE IT TEAM The IT team is faced with continuous challenges related to the acquisition of data from the shop-floor. The values associated with the capture of shop-floor data through Valor IoT Manufacturing include: • Reducing complexity of shop-floor data-acquisition software. • Eliminating need for normalization of data from different sources. • Managing data flow to avoid bottlenecks. • Reducing lead-time for IT software development and implementation. • Easier integration of shop-floor data for use in enterprise and cloud-based systems, including “big data” applications. FROM THE PERSPECTIVE OF GENERAL AND CORPORATE MANAGEMENT As well as the focus on individual production operations, companies with multiple production sites can combine data from Valor IoT Manufacturing to build a complete picture of the global operations, with the ability to compare sites in many ways. The benefits management can obtain from using Valor IoT Manufacturing include: • Direct operational visibility of all processes. • Visibility and comparison of key metrics, such as up/down/added value time, completions across all assets. • Understanding of individual process reliability and utilization, by equipment type, vendor, age, etc. • Visualization of trends and improvements. • Like for like performance and cost comparisons between equipment, lines, sites, products, customers, etc. Valor IoT Manufacturing—The Internet of Things for Electronics Manufacturing ©2016 Mentor Graphics Corporation, all rights reserved. This document contains information that is proprietary to Mentor Graphics Corporation and may be duplicated in whole or in part by the original recipient for internal business purposes only, provided that this entire notice appears in all copies. In accepting this document, the recipient agrees to make every reasonable effort to prevent unauthorized use of this information. All trademarks mentioned in this document are the trademarks of their respective owners. F o r t h e l a t e s t p r o d u c t i n f o r m a t i o n , c a l l u s o r v i s i t : w w w . m e n t o r . c o m 03-16 TECH14070 • Reporting of costs associated directly with manufacturing of specific products, end-to-end, for P&L analysis. • Reduction of any recall or rework through the use of traceability data. • Capacity analysis. • Reduced material related costs. • Reduced costs of poor quality. CONSULTANCY SERVICES We recommend working with the Mentor Graphics unique consultation services team to obtain the maximum potential value from the implementation of Valor IoT Manufacturing. They can assist with specification of the system, identification of potential applications, values and benefits, solution integration, and optimizing implementation. CONCLUSION Valor IoT Manufacturing represents one of the most significant steps forward in the electronics manufacturing industry, by directly enabling solutions that address critical trends of high-mix and the need for flexibility. As more of the industry demands computerization for intelligence that links automated processes together into a complete smart factory, Valor IoT Manufacturing provides the ideal environment for data acquisition and control. It also provides the architecture on which IT systems and computerizations of all forms can be based, using accurate, reliable, and timely shop-floor information. From the business perspective, the use of Valor IoT Manufacturing brings the shop-floor seamlessly through a revolution, without the need to “change your world.” Existing IT systems do not need to be extensively reengineered or production assets needlessly replaced. The values that can be achieved from the use of the data tip the balance in favor of local, automated, and flexible manufacturing, which serves to focus investment close to the market. They raise the level of competitiveness to the level at which local manufacturing can match, then exceed remote factories on a world-class basis.