AWS를 활용한 Digital Manufacturing 실현 방법 및 사례 소개 - Douglas Bellin, 월드와이드 제조 솔루션 담당 디렉터, AWS / 장대기 대리, GS Caltex :: AWS Summit Seoul 2019
4 likes•2,065 views
The document outlines various digital manufacturing use cases and technological solutions from AWS that enhance operational efficiency and decision-making in industries such as manufacturing, oil & gas, and healthcare. It details specific company profiles and challenges, showcasing implementations of AWS services for predictive maintenance, data analytics, and improved production outcomes. Key examples include products from Georgia Pacific and SKF, illustrating how AWS tools lead to increased profits, reduced downtime, and better customer service.
![© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Data Analytics 적용의 시작 그리고 Cloud
[생산 영역] 정형화된 Data 풍부
[생산 외 영역]
Planning
Scheduling
Optimization
Operation
Reliability/SHE
Sales/Marketing
Logistics/Procurement
Finance/Corporate Strategy
시범 과제
[알고리즘 활용]
Anomaly Detection
Cloud 도입의 필요성
• Pilot이었기 때문에 거대한
리소스를 필요 시에만 사용
• 공장 시스템에 영향을
주지 않아야 하는 제약
AWS Cloud 도입
• 다양한 가격 정책
• Cloud 대부분의 서비스가
Korea Region에서 운영
• 다양한 Data Analytics
Tool 보유
• 높은 보안 수준 수 천 개의 제어 장치와 수 만개의
유관 장치 테스트 시, 머신러닝을 위한
Computing 자원 필요
Optimization
4.52
4.53
4.53
4.54
4.54
4.55
현재 값 목표 값
Anomaly Point](https://image.slidesharecdn.com/day1track5session1-190418072124/85/AWS-Digital-Manufacturing-Douglas-Bellin-AWS-GS-Caltex-AWS-Summit-Seoul-2019-19-320.jpg)
AWS를 활용한 Digital Manufacturing 실현 방법 및 사례 소개 - Douglas Bellin, 월드와이드 제조 솔루션 담당 디렉터, AWS / 장대기 대리, GS Caltex :: AWS Summit Seoul 2019
- 1. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Digital Manufacturing and Relevant Use Cases Douglas Bellin, AWS 장대기, GS칼텍스
- 2. 발표자료 바로 공개 발표자료는 발표 종료 후 해당 사이트에서 바로 보실 수 있습니다 © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
- 3. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
- 4. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. REVENUE GROWTH OPERATIONAL OVERHEAD Empowered Sales Teams Increased Efficiency Intelligent Decision Making Products that Get Better with Time Better Relationship with Customers Data Driven Discipline
- 5. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Business Architecture – Manufacturing Operations PLC/ CONTROLLER IMAGE MES QUALITY MANAGEMENT SYSTEM HMI SCADA GW (OPC SRV) HISTORIAN ASSET MANAGEMENT SYSTEM Plant 1 Plant 2 Plant 3 Ingestion Batch Process Machine Learning Rule Engine Optimization Real Time Process Global Monitoring Dashboard Predictive Maintenance Quality Control Maintenance Management/ Operators Modeling Applications
- 6. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. AWS Manufacturing Reference Architecture Greengrass Edge/GW S3 Data Lake Kinesis MES Factory Machines ML Inference IoT Core Sage Maker ML QuickSight Business Intelligence Athena Historian Storage Gateway EMR EBS EC2 Batch AppStreamEBS EC2 E&D Workloads (PLM/HPC/CAE) Enterprise Workloads (SAP ERP/CRM)DMS RDS Local Servers RedShift Data Warehouse DataIngestion API SiteWise Snowball Edge Smart Products DynamoDB Lambda IoT Core Amazon Forecast Plant Maint. Planning Business Functions Greengrass Connectors IoT Analytics Timestream Outpost IoT Events EC2 Lambda Business Logic
- 7. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Use Cases for Industrial Data Platform Visibility Production Quality Access Critical plant information at any time Automate data capture Determine root cause Measure and improve OEE Drive continuous process improvement Reduce scrap and defects Compare KPIs and metrics across the Enterprise Predict out of control processes and failures Increase yield Benchmark machines, lines and plants Monitor regulatory info – track and trace, genealogy Control processes with stricter tolerances Measure and trend quality performance Increase capacity utilization Setup alerts to maintain process control Improve worker safety Reduce maintenance costs Reduce rework New services opportunities Increase in NPI time Smarter products
- 8. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Topics • Product Shape/Size/Color Recognition • Production Line Utilization (OEE) • Predictive Maintenance/Quality & Alarming
- 9. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Use-Case: Object Shape/Size/Color Detection Goal: Detect product shape/size/color, joints (weld/seam) problems as the product moves down a conveyor and/or (vehicle) assembly line, paint shop, etc. Industrial Camera NOTE: Pattern/Shape recognition is Machine Learning in it’s simplest form in seeing we have a defect but not really predicting why it occurred.
- 10. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Use-Case: Slow Machine Cycle Time Prediction Goal: Monitor a packaging machine for slow post cycle times, determine root causes/contributing factors (e.g., outgoing conveyor belt issues, etc.) then predict future slowdowns. PLC Packaging Machine PLC
- 11. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Paper & Building Product Manufacturing • Georgia Pacific is one of the world's largest manufacturers of tissue, paper, packaging, building products, and related chemicals. • The company employs over 35,000 people in more than 150 locations. • Products include brand names like Angel Soft, Quilted Northern, Brawny, Dixie, Mardi Gras, PerfecTouch, Sparkle, Blue Ribbon, Clutter Cutter, DensArmor Plus, DensDeck, DensGlass, DensShield, DryPly, FireGuard, GP Lam, Nautilus, Ply-Bead, Plytanium, Southern Gold, Sta-Strait, Thermostat, ToughRock, Wood I Beam, and XJ 85 Client Profile • Multiple facilities with different manufacturing equipment • Improving plant productivity & quality • A retiring workforce with valuable site-specific knowledge gained over years of experience • Downtime costs of $250k/day, and a need to maximize machine uptime • Amazon Kinesis to stream data from complex machines • Data Lake with Amazon Simple Storage Service (Amazon S3) to efficiently ingest and analyze structured and unstructured data at scale. • Amazon Elastic MapReduce (Amazon EMR) to transform data before delivering it in a structured fashion to data analysts through Amazon Redshift. • Amazon SageMaker, an AWS machine-learning solution to build, train, and deploy ML models at scale • 30 percent paper-product tear reduction. • Increased profits by millions of dollars for one production line, with hundreds of other production ines to benefit • Less dependence on disparate factory experts to keep lines running efficiently. • Predict equipment failure 60–90 days in advance, and reduce unplanned downtime. • Ensure the highest-quality product running at the fastest-possible rate Challenges Our Joint Solution Results “Overall, using AWS, we can ensure the highest-quality product running at the fastest-possible rate, so we can best serve our customers.” Steve Bakalar, VP of IT/digital transformation
- 12. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Bearing and Smart Product Manufacturing • Founded in Sweden in 1907 • The world's largest bearing manufacturer (they also manufacture seals, lubrication and smart lubrication systems, maintenance products, mechatronics products, power transmission products, and condition monitoring systems. • SKF employs 44,000 people in 108 manufacturing units. • Large industrial distributor network, with 17,000 distributor locations encompassing 130 countries. Client Profile • Move beyond selling only products, to a “Rotating Equipment Performance” service model • Ensure bearings are lubricated correctly and automatically to maximize performance and life • Gather data from customers all over the world to improve product design and performance • Add new replacement part revenue streams • Connected System 24 single point lubricator feeding a Data Lake with Amazon Simple Storage Service (Amazon S3) to efficiently ingest and analyze structured and unstructured data at scale. • AWS machine learning to help analyze products in the field • AWS Database services to manage large amounts of complex vibration and equipment condition data • AWS IoT services, and Lambda to speed time to market and lower costs on new product designs • Expand their revenue beyond a ship-and-forget model, to a services enhanced revenue stream • Grow sales, even if their raw product shipment numbers do not increase. • Innovate faster, with lower costs. • Concentrate on the value they can provide for their customers, instead of setting up the hardware and compute resources required to make it happen Challenges Our Joint Solution Results “I see a lot of speed of innovation coming from AWS, and we are confident that this is the platform we are going forward with.” Johan Tommervik, CIO
- 13. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Problem Wärtsilä needed to accurately predict, when the marine engines they manufactured needed to get serviced. Understanding and predicting the service schedule is vital for Wärtsilä to increase their service and parts revenue. Solution Accenture worked with AWS account SAs, AoD SAs, and Salesforce SAs to architect an IoT solution using Salesforce and AWS IoT Core to collect data and build predictive models. The solution developed is scalable and extensible beyond just this use case, as Wärtsilä has 14,000 ships with 35,000 engines installed. There are great possibilities for sensor driven IoT use cases. Impact The entire solution should result in an increase in parts/service sales for Wärtsilä and higher customer retention.
- 14. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Problem As Konecranes specializes in the manufacturing and service of cranes globally, they discovered that when they needed to make updates to their machinery it meant downtime and local presence onsite. Solution Using Greengrass has enabled them to deploy updates using cloud models that continually get smarter over time as they sync with the local environments. Impact This allows them to simplify their current crane architecture and make it possible to update calculations to the cranes in a secure way even after the installation has taken place.
- 15. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Problem Stanley Black and Decker finds it unsustainable to ingest, transmit, store, query and analyze all data generated at the edge and more specifically on construction sites or rural areas with constrained network resources. Solution AWS Greengrass enables Stanley Black and Decker to monitor and filter data at the edge of the network enabling applications to send asset health and predict any mechanical failures before they occur. Edge-based applications built on Greengrass will help detect and compare vibrations emitted by high value tools to historical signatures that indicate everything from normal operations to imminent failure. Impact Instead of trying to use all the data Stanley Black and Decker will utilize Greengrass to focus on the right data. Applications include remote troubleshooting of hydraulic assets by technicians, maintenance interval tracking, fuel savings, and alerts.
- 16. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. 장대기, 경영정보화팀
- 17. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Oil & Gas Industry, Refinery에 대한 이미지?
- 18. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. 정유업에서의 Data Analytics Energy Health care Logistics Automotive Consumer Packed goods Telecom, Insurance, Banking Retail Media Extremely limited Use of digital, primarily in internal operation 수 십 년간의 사업 운영을 통해 축적된 (Big) Data 오랜 시간 검증된 분석 솔루션 이미 Data Analytics를 적용 추가 개선이 가능한지에 대한 의문
- 19. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용의 시작 그리고 Cloud [생산 영역] 정형화된 Data 풍부 [생산 외 영역] Planning Scheduling Optimization Operation Reliability/SHE Sales/Marketing Logistics/Procurement Finance/Corporate Strategy 시범 과제 [알고리즘 활용] Anomaly Detection Cloud 도입의 필요성 • Pilot이었기 때문에 거대한 리소스를 필요 시에만 사용 • 공장 시스템에 영향을 주지 않아야 하는 제약 AWS Cloud 도입 • 다양한 가격 정책 • Cloud 대부분의 서비스가 Korea Region에서 운영 • 다양한 Data Analytics Tool 보유 • 높은 보안 수준 수 천 개의 제어 장치와 수 만개의 유관 장치 테스트 시, 머신러닝을 위한 Computing 자원 필요 Optimization 4.52 4.53 4.53 4.54 4.54 4.55 현재 값 목표 값 Anomaly Point
- 20. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용 영역의 확대 – 생산 영역 과제 발굴 영역 공정 수율/성상 예측 공정 운영 최적화 Trouble 원인 분석 이상 감지 우선 수행 과제 도출 업무 포털을 통한 예측 결과 조회 - 배치 분석 수행 및 결과 저장(DB) 업무포털 Cloud에 접속 후 분석 환경을 직접 활용 Use Case 예시 3 Use Case 예시 1 End-Point를 이용한 실시간 결과 조회 Use Case 예시 2 Inference (Response) Input Data (Request) 1) 분석 환경 구축 2) 모델 개발 “정형화된 분석” “계획적 반복 분석” “On-Demand 분석” “다수 Point에서 활용” “신규 모델 개발” “데이터 탐색 및 시각화”
- 21. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용 영역의 확대 – 고객 영역 ① 주유소 CCTV 영상을 통한 차량 인식 및 분석 예시 주유소 영상 차량 인식 모델 통계 Data 추출 및 분석 (방문 차량 모델 비율 등) Data Point 추출 (차종, 연식만) Privacy Data 원천 제외 고객 마케팅/주유소 운영전략
- 22. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용 영역의 확대 – 고객 영역 ① 주유 /세차 차량 인식 영상
- 23. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용 영역의 확대 – 고객 영역 ① 주유소 인근 교통량 인식 영상
- 24. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용 영역의 확대 – 고객 영역 ② 경쟁 주유소 파악 (예시) GS 칼텍스 주유소 주유소 Coverage (예시)
- 25. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics 적용 영역의 확대 – Risk Management 영역 100 110 120 130 140 150 160 170 180 p10 p50 p90 Actual Amazon Forecast <결과 예시> RM의 고민 • 급변하는 시장 상황 (변수의 증가) • 담당자 별 숙련도에 따른 예측 정확도 차이 다양한 변수를 고려한 방향성 예측 Logic이 필요 활용범위 확산검토 • 데이터만 잘 준비되면 코딩 없이 분석 가능 • Case 별 분석을 통해 영향력 있는 변수 산출 가능 비전문가도 활용 가능 확산 검토 중
- 26. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Data Analytics를 위한 AWS 활용 Industrial equipment Factory Data Interface Server S3 AURORA DB EBSEC2 Existing Analytics Tools (SAS) sas SageMaker EC2 기타 Data CRM CRM CRM
- 27. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Digitalization에 대응한 진화의 시작
- 28. 여러분의 피드백을 기다립니다! #AWSSummit 해시태그로 소셜미디어에 여러분의 행사소감을 올려주세요. AWS Summit Seoul 2019 모바일 앱과 QR코드를 통해 강연평가 및 설문조사에 참여하시고 재미있는 기념품을 받아가세요. 내년 Summit을 만들 여러분의 소중한 의견 부탁 드립니다. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
- 29. 감사합니다! © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.