IT機器・ITセキュリティ

2020年4月には、水産業に特化した初のリファレンス・アーキテクチャ(淡水、廃水、淡水化など)を監査し、認証を取得しました。ブループリントはシーメンスから提供され、国際的な工業規格に認定されています。

デジタルの発展はサイバーセキュリティの需要の高さを意味し、シーメンスは長い間多くのユーザーや企業に垣根のないサイバーセキュリティサービスを提供してきました。しかし、急激なデジタル化の発展により、ニーズの需要が高まり、立法者側からサービス運営側のセキュリティ強化を求められました。水産業においてサイバーセキュリティがこのように大規模に取り上げられたケースは過去になく、適合するために独自の新基準が必要であり、新しいプログラムを開発しなければなりませんでした。

試行錯誤の末に、新しく作られた当該認証は、シーメンスのプロセスとテクニカルリファレンスアーキテクチャの双方に規格のセキュリティ要件に準拠しており、シーメンス向けにテュフズードからIEC 62443認証は得たのは水産業では初めての快挙となります。

世の中のデジタル化は、私たちの仕事、生活に様々な変化をもたらしております。利便性がます一方で、私たちのデータやデジタルインフラのセキュリティ、プライバシーの安全を確保することが重要です。当社では、産業界のデジタルトラストのリーダーとして、サイバーセキュリティーに対する意識を高めることを目指しています。

私たちは、信頼を醸成するというテュフズードの使命に従って、信頼憲章の一員であることを誇りに思っています。グローバルなサイバー・セキュリティー・アライアンスは、2018年にミュンヘン安全保障会議でシーメンスによって設立されました。17の所属メンバーは、技術、航空、製造業など最新技術を駆使したグローバル・リーディングカンパニーです。この率先は他近隣国と共に、最先端のテクノロジーを備えた有力なエキスパートを集約し、世界中の資源を追求するものです。

私たち、テュフズードは自主性、専門領域、規制を通じて、重要なシステムや機密情報を保護し、お客様の事業サービスのサポートすることで、信頼憲章と１０の原則に貢献することを目指しています。当社のサービスポートフォリオは、テスト、監査、検査、認証、トレーニングを含むコアビジネスの中で、サイバーセキュリティは重要な部分を占めております。

Charter of Trustの詳細はこちら

Lifts perform a heavy duty – all day, every day. To predict optimal maintenance and prevent breakdowns at an early stage, TÜV SÜD Digital Service dove deep into the possibilities of the Internet of Things (IoT) and smart algorithms and developed an innovative solution. In Singapore, our team successfully implemented a smart lift testbed project for 118 lifts.

Tapping on their profound base of expertise in this field, our team designed and implemented tailormade algorithms for 24/7 operations monitoring, recording and anomaly detection based on sensor data. In a joint venture of lift experts and a sensor hardware provider, we developed a centralized, cloud based IoT solution to process almost realtime data on the physical condition, utilization, ride comfort, and anomaly detection of lifts. To make things even more demanding, the solution had to be applicable to different lifts, independent of the Original Equipment Manufacturer (OEM), age and type.

The project then moved into the handson phase: We equipped several lifts with laser and multi sensors on top of the cabin, which provide detailed information on the acceleration, position in the shaft, and more. Environmental variables such as pressure, temperature, and humidity are monitored as well.

The project proved to be an overall success: The testbed had a resounding 85% accuracy in failure prediction with the innovative analytics engine. The information collected provides a basis for automated predictive maintenance which will reduce costs and down time. And the story goes on: TÜV SÜD Digital Service already won another pilot project for implementing the innovative lift predictive maintenance solution on high-rise and industrial properties.

Making business fit for the future with Industry 4.0? Can do. For a leading industrial packaging product manufacturer TÜV SÜD Digital Service in Singapore developed a strategic roadmap and corresponding business case to create the path for Industry 4.0 transformation.

Guided by the vision and future targets of the client, the TÜV SÜD team worked to conceptualize and prioritize the I4.0 technology solutions that would be needed to realize the future state.

In a nine-week engagement we got busy to make the vision become reality. The current state of a factory was assessed, and our experts identified development opportunities in automation, connectivity, data analytics, as well as in the underlying IT and OT system architecture. Further analysis was conducted on the enterprise systems, databases, operations management, and strategic imperatives.

With defined transformational direction and targets, our experts crafted a high-level roadmap, and established the strategic priorities, investment requirements and cost benefit analysis. The Return on Investment (ROI) was projected for the entire Transformation program.

The final deliverables in the analysis and recommendations served as a basis for investment decisions of senior management at the global Head Office, direct the transformation journey at the local pilot site and serve as a blueprint for other plants.

In February 2021 we kicked off Drivers Club, an app by Division Mobility for their driver’s license applicants, with a remote design sprint.

The goal of the design sprint was to further develop the product within a short time span of three weeks, with the support of user input, by building a visual concept for user guidance.

By starting the design sprint with expert interviews, we collected as much relevant information as possible from every expert field, which also enabled us to scope out the project. Based on this input, all participants defined different features of the product and developed the user flows. These resources gave us enough foundation to start visual thinking, asking every participant to scribble suggestions for the visual implementation of the defined requirements. With these conceptual ideas in place, the whole design sprint team defined the direction for the visual prototype under the eye of the involved UX department. On the one hand, the resulting prototype was utilised for user testing, which enabled us to evaluate the product ideas with real users and therefore reduce the risk of expensive development errors.

On the other hand, it represented a solid foundation on which to define the requirements for a product that users will love.

At TÜV SÜD, we developed an AI-based solution to detect damage such as scratches and dents on cars. Pictures are taken of all sides of the car and then automatically analyzed to generate damage reports. As the setup works in near-real-time, the hardware, software, and architecture could be continuously improved, making the algorithms and services we employ as efficient as possible.

Our data engineers and cloud architects were responsible for designing, setting up, and constantly improving automated machine learning pipelines and a scalable cloud architecture that we used for data management. The key task for our data science team was to select, train, and retrain deep learning algorithms. As a result, we reached a high damage-detecting accuracy using region-based convolutional neural networks (R-CNN).

An active-learning approach also allowed us to integrate feedback on model retraining from our car experts to continuously improve the system. With the AI-based solution, we support our car experts and speed up the whole process.

This project is an excellent example of bringing the latest deep-learning-based approaches to life by integrating them into a scalable system for automated damage detection.