부산시청 도서요약
국내도서 요약
국내도서 프리뷰 해외도서 프리뷰
글로벌 트렌드 미디어 브리핑스
   글로벌 트렌드 

책표지








  • When Drawings Fly and Parts Are Born On-Site

    - Certified Additive Manufacturing Moves Beyond Prototypes into Aerospace, Defense, and Supply Chains

    The next market for 3D printing is not a model sitting on a desk but a part flying in the sky. Producing discontinued aircraft components and urgent defense spares where they are needed requires a demanding certification system to move before the printer does. Manufacturing competition is shifting from supply chains that transport finished goods to supply chains that transmit verified designs and processes.

    [Key Message]
    * The industrialization of additive manufacturing depends more on repeatable quality than on printer performance. Certification begins with evidence that the same design produces the same component across different machines and locations.

    * The first aerospace and defense market is closer to discontinued parts and urgent maintenance than to mass production. The economics of digital inventory and on-demand production become more compelling when low-volume parts are difficult to reproduce with new tooling.

    * Distributed production is not a matter of emailing a file and printing it locally. Approved designs, materials, machine condition, operator qualifications, post-processing, and inspection records must form one continuous digital flow.

    * Certification costs can destroy the business case if every component starts from zero. The key to expansion is a qualification framework that validates processes, machines, and material families and reuses the evidence for similar parts.

    * The opportunity for Korean companies extends beyond printer sales. A certified production platform combining material data, process monitoring, nondestructive inspection, and digital traceability can create a much larger market.

    ***

    From Prototype to Flight-Approved Part
    An aircraft sits idle in a maintenance hangar. The failure is minor, but one metal bracket is missing. The original supplier has stopped making it, and no inventory remains. Under the traditional approach, a new mold must be created or a production line reopened to satisfy a minimum order. Restarting production costs more than the part itself, while the aircraft remains grounded for weeks or months.

    This is the moment when additive manufacturing looks attractive. A digital design is loaded, a thin layer of metal powder is spread, and a laser melts only the required area. Repeating that process thousands of times creates a complex component. Unlike machining, there is no need to cut most of the material away from a large block, and no casting mold is required. Setup costs can fall sharply when demand is measured in ones or tens.

    Yet matching the shape does not make an airworthy part. A desktop prototype can simply be printed again if its dimensions are slightly wrong, but a component near an engine, wing, or landing gear must endure fatigue, vibration, heat, and corrosion for years. A microscopic pore or incomplete bond between layers can become a crack after thousands of flights. The aerospace question is not “Can it be printed?” but “Can we prove that it was printed with the same performance every time?”

    In 2023, the U.S. Federal Aviation Administration issued guidance for demonstrating compliance of powder-bed-fusion additive manufacturing processes used for aircraft engine parts. The European Union Aviation Safety Agency has also continued developing guidance for applying metallic and nonmetallic additive manufacturing across aircraft, rotorcraft, engines, and auxiliary power units. These developments signal that additive manufacturing is moving from an exceptional experiment into a manufacturing process governed within established airworthiness systems.

    A Manufacturing Chain Longer Than the Printer
    The quality of an additively manufactured part is determined long before the print button is pressed. Changes in particle size, shape, oxygen content, and the reuse history of metal powder alter how it melts. Gas flow inside the machine, laser power, scan speed, and build-plate temperature also affect the microstructure. Even on the same machine, residual stress and strength can vary according to the position and orientation of the part.

    When printing ends, roughly half of the manufacturing work may remain. The component must be separated from the build plate, supports removed, and heat treatment and surface finishing completed. Computed tomography, ultrasonic inspection, dimensional checks, and material tests may be required to find internal defects. An aerospace component is not merely an object made by a printer. It is the outcome of a complete process extending from powder control to post-processing and inspection.

    Certification authorities therefore examine more than one finished item. They assess design allowances, material specifications, machine performance, process variables, operator qualifications, inspection methods, and recordkeeping. The joint ISO and ASTM standard for performance and reliability testing of laser metal powder-bed-fusion machines for aerospace applications points in the same direction. Criteria are needed not only at initial installation but also after relocation, repair, or major changes.

    Transferring production between factories is not completed with a copy command. Even identical printer models from the same manufacturer can differ because of installation conditions, machine history, actual laser output, and the condition of the powder-delivery system. Moving an approved process from one site to another requires reference specimens, comparisons of density, strength, and surface condition, and evidence that machine-to-machine variation remains within an accepted range. The real unit of distributed production is therefore not a printer but a qualified combination of machine, material, operator, and inspection system. Only while that combination remains controlled does moving a design file also move manufacturing capability.

    This creates the paradox of additive manufacturing. It appears to remove tooling and simplify production, yet industrialization increases the volume of data and quality control. Traditional manufacturing relies on material specifications and process experience accumulated over decades. Additive manufacturing generates new data about lasers, melt pools, temperature, and oxygen concentration while a component is being built. More sensors will not accelerate certification unless those data can be converted into evidence of quality.

    Discontinued Parts Open the First Market
    Additive manufacturing does not make every component cheaper than established methods. Casting, forging, or machining may still be superior when hundreds of thousands of identical parts are required. Printers have limited build rates and must carry the cost of expensive powder, post-processing, and inspection. The first market therefore appears where demand is low, procurement is difficult, and the cost of a stockout is high.

    Satair, an Airbus services company, demonstrated this logic when it converted an A320ceo wingtip-fence component to additive manufacturing. The cast part was no longer available from the original supplier, and recreating tooling for a small order did not make economic sense. Satair converted the component to titanium additive manufacturing, completed a new certification process, and supplied it with an EASA Form 1. Recertification took about five weeks, while four components could be printed together in a 26-hour build. The company reported a 45 percent reduction in nonrecurring costs compared with conventional alternatives.

    The important change is not printing speed but the definition of inventory. In the past, companies forecast demand and stacked physical parts in warehouses. In the emerging model, approved design files, process conditions, and inspection plans are stored as digital inventory, and production begins at a qualified facility when an order arrives. What is stored is not simply a part but the authority and evidence required to manufacture it.

    This model is particularly valuable for long-lived assets such as aging aircraft, ships, and power equipment. Equipment operated for decades often outlives its original supplier, loses complete drawings, or faces excessive minimum order quantities. Three-dimensional scanning and reverse engineering can restore geometry, but copying shape alone does not recover design intent or material performance. The required loads and failure modes must also be understood before the result becomes a certified component.

    Economics should not be calculated from unit price alone. A conventionally manufactured item may cost less per piece, yet additive manufacturing can become preferable after the cost of recreating tooling, minimum orders, long-term storage, obsolete inventory, and operational downtime is included. Conversely, an item replaced frequently with stable demand may remain better suited to an established production line. The key question in selecting additive-manufacturing candidates is not whether a part can be printed, but how much total cost created by procurement failure can be removed.

    Production Moves to the Front Line and Maintenance Depot
    In defense, a delayed part is not merely a cost problem; it can limit mission capability. When an aircraft or ship deployed far from home stops because of one small component, crews may wait for delivery from a central warehouse. Additive manufacturing offers the possibility of producing closer to the point of need. Yet the military does not seek freedom to print anything quickly. It needs the ability to reproduce approved parts safely.

    In 2026, the U.S. Navy's Fleet Readiness Center East announced its first delivery of flight-worthy metal additively manufactured parts to the fleet. They included a weapons-pylon fitting for the AH-1Z Viper, a main-landing-gear repair fitting for the V-22 Osprey, and a blanking plate for the C-130 Hercules. The depot completed a capability demonstration for its equipment and processes in less than six months and received formal validation that its printed components met the same safety and quality requirements as conventionally manufactured parts.

    The case shows a realistic sequence for distributed production. Organizations do not begin by printing the most critical rotating engine component at a forward location. They start with tools and flight-worthy parts whose risks are comparatively manageable, building trust in machines, operators, and inspection systems. Materials and component criticality then expand step by step. Deploying printers takes less time than expanding the approved production envelope.

    Distributed production also does not eliminate central control. Instead, the network must manage which facility is qualified to produce which component using which machine and material. A print order should not be released if machine calibration has expired or the powder history is unclear. As printers spread across battlefields, ports, and maintenance depots, central digital authorization and local physical quality assurance must become more tightly connected.

    Trust Is the Real Asset in Digital Inventory
    Design files are easy to copy. That is both the advantage and the risk of distributed manufacturing. An unapproved version may be printed, a file may be altered, or an unauthorized supplier may produce a component with the same appearance. Moving design intellectual property and export-controlled information across networks also creates new cybersecurity concerns.

    Digital inventory therefore must contain more than geometry. The design revision, permitted material, build orientation, machine-setting ranges, post-processing procedures, inspection criteria, and approved production sites must travel together. The system must track who accessed the file, when and where it was printed, which powder batch was used, and what the inspection results showed. Each component needs a digital identity connecting its manufacturing history to its physical serial number.

    In-process monitoring can strengthen that trust. Cameras and sensors can observe powder layers and melt-pool conditions in real time, identifying warning signs earlier. But the existence of abundant sensor data does not prove the absence of defects. The relationship between a signal and an actual pore or crack must be validated, and the same judgment must work across machines and sites. This is why aviation authorities continue to examine fatigue, damage tolerance, nondestructive inspection, and in-situ process monitoring.

    Artificial intelligence can compare large volumes of build data, find unusual patterns, and predict quality. In certification, however, explainability and change control matter alongside accuracy. Organizations must determine whether an approved envelope remains valid when a model is updated and who is responsible for a false acceptance. Digital transformation in manufacturing is less about eliminating paper certificates than about preserving the evidence behind every decision for longer.

    The Competition Is to Sell Certified Production Networks
    The largest bottleneck to additive-manufacturing expansion is not a shortage of machines but the cost of qualification. If material testing, fatigue testing, and process validation begin from zero for every component, the advantage of low-volume production disappears. One emerging solution is to validate a common range of materials, machines, and processes first, then reuse that evidence for families of similar parts. The goal is to turn trust in the production system itself into an asset that extends beyond a single approval.

    Corporate adoption should also proceed in stages. First, identify parts with long lead times, high obsolescence risk, and costly failures. Second, rank candidates by safety criticality, geometric complexity, expected demand, and existing tooling costs. Third, produce through qualified service providers while accumulating material and inspection data. Fourth, establish certified production capability internally or at regional hubs only after sufficient part families and repeat demand are confirmed. Buying a printer first and then looking for something to print is the sequence most likely to fail.

    Korea combines capabilities in aerospace, defense, shipbuilding, nuclear power, automobiles, metal powders, and precision machining. Equipment localization alone, however, will not overcome the high barriers of global markets. Customers do not buy laser output; they buy an approved part and shorter maintenance downtime. Material databases, process sensors, nondestructive inspection, secure design repositories, and automated certification documentation must be bundled into one service.

    3D printing will not eliminate supply chains. It creates a new supply chain of feedstock powder, heat treatment, inspection equipment, and certification expertise. What changes is the object being transported. Instead of shipping every component from a distant warehouse, companies transmit verified designs and manufacturing conditions and convert them into physical parts where they are needed. The strongest future supply chain will not be the network with the most printers. It will be the network that produces the same quality at every location and proves that fact quickly.

    Reference
    Federal Aviation Administration, June 2023, AC 33.15-3: Powder Bed Fusion Additive Manufacturing Process for Aircraft Engine Parts
    European Union Aviation Safety Agency, September 2025, CM-S-008 Issue 3: Additive Manufacturing
    ASTM International, November 2020, ISO/ASTM 52941: Acceptance Tests for Laser Metal Powder-Bed Fusion Machines for Metallic Materials for Aerospace Application
    Airbus Satair, October 2020, Satair Provided Metal Printed Flying Spare Part
    United States Navy, April 2026, FRCE Delivers First Metal 3D Printed Parts to Fleet