Nokia D.A.R.T focus in Manufacturing
From Grant Marshall, Head of Supply Network and Engineering, Nokia
Factories have been around for a long time and have evolved greatly through the industrial revolutions. The latest industrial revolution (Industry 4.0) is definitely making significant changes. 3D Printing, Automation Robots, Digitalization and Connectivity are just some of the concepts being developed that change manufacturing as we know it.
To enter into Industry 4.0, Nokia developed the Conscious Supply Network concept that provides a wider view than only manufacturing concepts by I4.0, as it focusses on overall supply chain 4.0. Nokia translated these concepts into practical application that we call our Conscious Factory. The Conscious Factory fits into an end to end supply network that provide full transparency to each step in the supply chain process.
Key technical trends that will impact electronics manufacturing
Let’s break the Conscious Factory into a few strategic areas: Digitalization, Analytics, Robotics, and Transparent (DART).
Before we get into specifics on each area, let’s first discuss potential technology trends that will also impact manufacturing strategies.
Additive Manufacturing
One technology trend showing significant impact is the 3D Additive Manufacturing (AM). Not new, but becoming more and more footprint and cost beneficial.
3D AM by definition supports digitalization of the mechanical manufacturing arena. Machine functions, build accuracy, build speed or building room space are constantly improving.
We must be aware that what becomes possible today will become standard some months later. New materials and the freedoms of designing do the rest. Global spare part management become easy, you do not need huge warehouses. Instead you build where needed by selecting 3D printing farms over the globe, saving time, transportation costs and carbon footprint making solution more green. The economic approach of 3D use is increasing on a daily basis compared to conventional manufacturing approaches.
Close attention must be planned to accommodate 3D printing strategies. What can be 3D printed and what will still require conventional approaches, simply due to economical reason? The question of quality insurance will need a field of innovations. Component traceability, field retrofit exposures, quality printing farms, 3D printing raw material etc. all need innovative solutions.
Big Aerospace companies have already proven some success in much of these areas. They have already released the process for use to build bionic and ultra-lightweight products.
New materials and processing methods
In electronic manufacturing, new materials and new ways for processing the manufacturing will start to shape product designs.
We see here the use of LDS (laser direct structuring) that allows the use of 3 dimensions to make electronic circuits. The New 3D circuit with support of active electronics can change the principles of product design in such a way that will result in less number of PCBs (printed circuit boards). This will create a more economical usage of the footprint, and supports the slogan make products smart and intelligent.
The printed electronics is another challenging technology in this area. In combination with alternative materials we can use this for new designs and additional functionalities. The ability to make multi dimension circuits will change product designs saving footprint and ultimately costs of transportation.
The next step of optimizing footprint is the embedding of components in between PCB layers. This also can be used for footprint reduction and implementing additional new capabilities.
Component interconnect/bonding strategies
Component interconnect/bonding strategies is the next are of interest. The traditional process using solder paste has limitations with extreme thermals stress factor and cannot be easily adapted to new extreme temperature sensitive materials and 3D structures. Here new conductive glue and nanofoil approaches can be used.
The use Conductive Glue can reduce the thermal impact of existing manufacturing process for the applications that can’t handle the existing temperatures used in the existing soldering process. Alternate activators can be used to achieve the needed results.
Nanofoil – functionally proven technology today however not yet cost effective in manufacturing process compared to existing solder paste. But the idea here is by placing a small amount of Nanofoil on the soldering pad and placing a component on it, then applying a small current or laser to the nanofoil, the nanofoils heats up to extremely high temperatures for Nano seconds, melts, and then you are left with a perfectly ultra-thin bonded component to a PCB with nearly no solder void problems of today’s processes.
Track and trace solutions
Track and trace solutions are critical for the conscious supply network. Each type of indoor and outdoor location tracking solutions – regardless of final technology chosen, Bluetooth, RFID Ultra-wideband or narrow band LTE, 5G solutions – needs to provide a secured and complete visibility into the supply chain. It has to understand all material arrival schedules, manufacturing build schedules, and delivery schedules in order to meet customer expectation or dynamically change distribution strategies to recover from delays in supply network.
Connectivity
Although not a manufacturing technology, connectivity solutions will impact manufacturing process greatly.
Today’s factories are hard wired with tens of kilometers of cables connecting the machines. These hard wires make factories somewhat inflexible. Each layout change creates the need for rearrangement of the IT infrastructure. Existing wireless technologies like Bluetooth and Wi-Fi simply have too many limitation to overcome.
New possibilities given by the rollout of LTE and 5G spectrum provide businesses extremely secured own private networks to connect all equipment via dongles with the network access and maintain full internal IT support.
Factory connectivity will also change with the introduction of hundreds if not thousands of IoT devices and sensors, all sending and receiving data. With more video based application being enabled, higher bandwidth, security and lower latency solutions are critical.
Service organizations need to remotely participate in troubleshooting activities and quality monitoring. Moving to wireless provides footprint reduction on shop floor, power reduction for operations, and security improvements.
AR and VR
Finally Augmented Reality (AR) and Virtual Reality (VR) are really starting to hit their stride. VR solutions are excellent methods to help with training requirements of new hires, but do not show benefit over AR solutions on factory floors today. With AR technology, regardless of whether it comes in handheld solutions or wearable solutions, the technology interacts better with the environment.
AR Technology can be used in three main ways:
- In terms on displaying work instructions for an operator. Using the Augmented reality to overlay the work instructions or process steps on top of the physical environment to make sure everything is done correctly
- Picking selections and highlighting what to grab and where to look
- Most important monitoring quality control with video recording capabilities of the AR devices coupled with Edge computing real time video analytics can be established to guaranty all work instructions were followed and output quality is 100%. This video analytics requires a very strong high bandwidth low latency secure network.
These AR/VR solution will allow for very fast onboarding of new resources and very fast movement of manufacturing knowledge from one site to another. This basically mimics the behavior of a Hardware Robot where you transfer a software program from robot in site 1 to a robot in site 2 with a push of a button. AR solutions can create similar concept for work instructions, visual cues and automated quality control. Achieving a “Push Button Product Transfer” utilizing resources that can be on boarded in minutes instead of months while still maintaining the necessary quality.
I have only mentioned a few technologies that will change the future of electronic manufacturing industry. If some or all of them will turn into reality in the next 5 years I see a point where many smaller factories would popup satisfying region specific needs and offering competitive services. However before this could be realized, several of the main industry 4.0 trends need to also align.
Nokia’s Conscious Factory value proposition
Industry 4.0 DART key strategic areas, when coupled to the technology evolution mentioned above, deliver Nokia’s Conscious factory value proposition:
Digitalization
Everyone has heard of digitalization. It’s a statement easily made but harder to achieve.
Fundamentally in its simplest form, it is a concept that everything is electrically connected not needing human translations. E.g. plans are transferred via systems, not emails, machines are real time reporting their sensor data to show progress and quality.
Digitalization must be developed to interconnect everything, leverage cloud technologies, and allow for real time understanding from a distance. With Digitalization comes a new level of transparency. End-to-end supply chain flows can become visible to needed teams.
Analytics
Analytics is much more than an 9 letter word that means graphs. Graphing data is just visualization of data, it is not analytics. Analytics is taking data from many sources understanding correlation and causation analysis to be able to develop prediction and prevention strategies. Visually showing that variable 1 grows so output shrinks shows correlation not causation. The output and the first variable may both be changing as a result of other variables. Causation is key to root cause.
Machine data is just one format. Analytics must be able to take data in many forms, structured (Integers, Real numbers, Boolean values like in a databases) and merging in un-structured data (Heat, Time of day, cafeteria menu, historical email trails, voice recordings, etc.) and look for patterns.
Develop causation and prediction models is the end state of Analytics. Typically the journey is 5 steps before true analytics is possible:
Robotics
Robotics and automation are closely connected. Robotics is not only the use of 6-axis hardware, it can be everything from Robotic Process Automation (RPA), to the factory hardware robots, autonomous driving pick-move-place applications like Drones, and much more.
RPA is a software program that copies a repetitive step that a human would do on a computer, and automates it. Good example would be to look through emails for a specific type and process the attachments into a database. This concept frees the individuals from the “copy and paste” work to true data analysis and value add work.
Hardware Robots are used in factories. Both collaborative and non-collaborative robots exist in the industry. Low volume and high mix products with shorter lifecycles make automation by robots extremely complex and costs intensive, thus making the investment in hardware robotics harder. Newest technologies of computing autonomous systems and easy changeover between the products open new chances to get our products automated by robots and can generate higher ROI on the investments.
Finally Drones – in this category I merge the traditional Helicopter like drone flying in the sky and the land based Automated Guided Vehicle (AGV) and Autonomous Indoor Vehicle (AIV).
In factories, the need to move large heavy components will be around for some time. Simply put, flying drones would not be best suited for this. But we all agree that having people push heavy carts of raw material around is something a Drone can do better.
Initial Drones were AGV that followed a magnetic tape on the floor to their destinations. If the tape got damaged the drone stopped working. The next generation of land based delivery drone is the AIV come in two main ability levels:
- Follow a specific person with self-driving awareness of position and auto drive to specific drop off locations,
- Automatically learn as you go. Simply give it a destination and it will learn and memorize the best route and be intelligent enough to redirect if a blockage occurs much like car navigation systems do today.
Regardless of the robotic solution, all types of robots are cost effective today and it is a site by site based experience on where to invest first.
Transparency
Transparency of data from many sources. Bringing together field data, performance data, supplier data, R&D data, manufacturing process data into a common system that can be leveraged to provide full analytics on end-to-end supply chain to help business optimize and meet customer expectations, and give customers needed transparency into their deliveries, updates on arrival times, and early knowledge of situations that are impacting their deliveries.
Transparency for customers should be almost app like where they can see the truck delivery in real time, have early warning of its arrival to ensure necessary people at receiving dock to receive goods and process for deployment as quickly as possible.
Nokia's Conscious Factory – a factory of the future, available today
Nokia’s Vision of the Conscious Factory is an agile and smart Manufacturing Service, fully automated and green that is self-learning and able to predict and prevent: “The Conscious Factory!”
With all the above technology changes, we believe the concept of large manufacturing locations will also start to change. Smaller agile factories could pop up to better meet regional requirements and if the interconnectivity and technical ramp up challenges are solved with Cloud and AR type solutions factory count is no longer a concern in the decision making.
Conscious Factory-in-a-Box
Nokia has developed a specific solution for this that we call our Conscious Factory-in-a-Box (FiaB). FiaB is simply taking our existing Conscious Factory and developing a “Lego” building block factory using standard Cargo containers.
These Lego factories are much better suited to meet regional and innovation startup requirements. They can be as large as needed to meet the build requirements by simply adding additional containers to the location.
Typical applications for FiaB would be:
- Meet specific country of origin requirements – Drive the factory to an open field, solar powered, and find workforce, provide AR/VR solution, divert material to a GPS Location, then build product with guaranteed process adherence and quality using video analytics.
- Provide New Product Introduction (NPI) services – With many startups, hardware manufacturing takes time and usually involves travel. By physically moving the factory to the R&D location to allow for fast design iterations and prototype builds, the time to market can be enhanced giving competitive advantage
- Training and Repair services – Factories are constantly in hiring phases, and part of the hiring process is training. With the technology advancements and the FiaB solution a small separate factory can be created dedicated for large scale training without impact overall site production.
- Factory Out of the Box – As individual sites experience capacity constraints, a container based factory can be shipped to location to serve as relief capacity. It can be operated and managed from within the container in situations like Disaster recover mode, or simply pull the equipment from the movable pallets inside the container to the shop floor to provide additional lines needed to meet temporary peeks in capacity and eliminate the bottlenecks.
Conscious Factory and the Lean Principles
Factory people are usually very familiar with LEAN principles. Nokia’s Conscious Factory in conjunction with the conscious Supply network help address all of the well-known “7 wastes” to avoid:
- Overproduction – Many products today are made up from components with various lead times from days to months, and customers have changed their behaviors and are starting to ask and expect almost immediate delivery of their product in a “Just in Time” concept. Supply Chains have been forced to build inventory to meet agility expectations of customers. This inventory is “overproduction” because it is built without a known confirmation of orders. With Nokia’s DART solution and SMART products providing information about in field utilization and performance, Nokia’s Supply chain can add this information into its analytics and predictive planning process to ensure we have necessary inventory at the right time to build the right products to meet customer expectations.
- Waiting – With robots in factory many of the handoff points between stations can now be automated where previously manual lines used “Work in Progress” (WIP) buffer stock. The removal of many of these buffers can be achieved with automation. This will drive to a closer single piece flow model.
- Transporting – FiaB is a great solution to build in transit. Most raw material comes from China and is shipped to factories around the globe. For Europe and Americas typical transit is boat where the raw material sites in cargo containers for 3 weeks then is sent to a factory, unpacked, received, built into finished goods, repacked, loaded on trucks and sent to customers (Usually taking 2-3 weeks more time). The future state is with FiaB. While the raw material is on the boat it is built into finished goods and when it arrives at the destination it is delivered directly to customers, savings weeks from the overall lead times and eliminating transportation waste.
- Inappropriate Processing – Too much process steps can be avoided on more automated and digital solutions. An example would be manual inspections after a step. With AR for step by step work instructions and Video analytics monitoring for adherence and quality, post process inspection becomes redundant.
- Unnecessary Inventory – like description of Overproduction and Waiting. Nokia’s Conscious factory working with DART principles and SMART products clear drive better forecasts, thus reduced inventory, and Automation strategies and elimination of many of the WIP points.
- Unnecessary / Excess Motion – When bottlenecks exist in manufacturing process, WIP is built up, and must be moved around for storage the first form of excess motion. Next would be manufacturing steps can not automatically detect errors happening in real time preventing them from getting the next steps. With the Nokia’s Conscious factory, AR and video analytics error detection is immediate prevent errors from moving forward to next steps and saving unnecessary movement backwards for rework
- Defects – As described in Excess motion above, Nokia’s Conscious factory leveraging the new technologies like AR/VR and Video analytics drive earlier defect detect and prevention strategies. Preventing errors from propagating in the line causing more costly rework or worse scrap.
Nokia’s Conscious Factory brings everything together
Starting from the right-hand side of the above graphic, products we sell and build will become smarter. This means the product needs a level of self-awareness to report back key vital information like usage and critical failures in real time like with new advanced flight recorders. With this knowledge, Nokia end to end supply chain can better plan with its suppliers upcoming material requirements to meet customer expectations.
Factories can be large physical campus, or small Cargo Containers. With Digitalization and cloud technologies number of locations is less important than ideal location to meet customer expectation.
Cloud solutions are critical to success. Cloud to MES (Manufacturing Execution System) and ERP (Enterprise Resource Planning) is a given, but even for testing. Test assets must move to cloud in order to provide full flexibility. Cloud based testers can be better utilized reducing CAPEX investment needs. Test assets also can have long lead times. By having testers in the cloud and with AR technology a new site can be enabled with much shorter lead time
AR and VR technology will heavily be used to ensure quality and fast onboarding or new products to a location or new people at a location. As well as provide platform for remote supervision and support collaboration environment for faster access to technical experts.
Automated factories can run with remote supervision. People can be dispatched as necessary for maintenance or troubleshooting, where again AR technology will guide operators through the process and physically to the location needing maintenance.
Sensors and location tracking with IoT power and lighting control will be used internally to guide people to desired location. Sensors data can feed the big data analytics engine to identify and recommend changes to improve performance and determine maintenance scheduling. Sensors attached to drones and AIVs can be used to automatically open needed doors for material movement.
Robots: Robots for assembly, robots for pickup and delivery and robots for packaging. And of course the robots need sensors to collect information to the analytics. Hardware and software robots work collaboratively with people to achieve results.
IoT: Internet of Things is a term for devices that get connected. Sensors, machines, handheld devices, Drones, Auto Power systems are examples of IoT that need to work collaboratively in the factory environment. Intelligent power control systems can save significant energy costs by ramping down equipment that is idle, turning lights off where people are not physically.
Analytics: Analytics is key for so many items. Material planning, Capacity planning, Maintenance scheduling, and overall site effectiveness. But Analytics with digitalization will be limited. It’s critical they get developed together.
End to End (E2E) supply chain tracking will be achieved. From supplier through to customer. Automated Good received and on hand inventory and goods shipped will be achieved my merging outdoor track and trace solutions with indoor positioning solution, and integrated with customer locations for their automated Goods receiving.
It is capable today, and Nokia’s bringing it all together, making it happen.