Hinweis: Die aktuelle OOP-Konferenz finden Sie hier!
SIGS DATACOM Fachinformationen für IT-Professionals

Die Konferenz für Software-Architekturen
30. Januar - 03. Februar 2017


Vortrag: Di 3.3
Datum: Di, 27.10.2020
Uhrzeit: 14:45 - 16:15

Quality Whisperer - Self-learning AI Improves Production Quality in Complex Variant Processing

Uhrzeit: 14:45 - 15:30
Vortrag: Di 3.3 1)


ZF plant Saarbrücken manufactures around 11,000 transmissions per day with 700 variants. Process experts from different company areas had to spend a lot of time to discover influencing factors for root cause of minor product quality.

Therefore, an AI project was started with company IS predict with the objective to get reliable and fast results on root cause discovery. This saves time and reduces minor quality significantly. Complex root-cause findings can be reduced from several days to hours.

Target Audience: Responsible for production and / or quality, technical plant managers, CDO, innovation managers
Prerequisites: None
Level: Basic

Extended Abstract
The ZF plant Saarbrücken, Germany, manufactures around 11,000 transmissions per day. With 17 basic transmission types in 700 variants, the plant manages a large number of variants. Every transmission consists of up to 600 parts. Each transmission is 100% tested in every technical detail before shipment. The plant Saarbrücken is a forerunner and lead plant in innovative Industry 4.0 technologies. Therefore, activities were started to tackle one significant challenge, which is caused by the enormous variant diversity: Finding root-causes for unsuccessful end of line testing. The management of the complexity is a big challenge because transmission parts can be produced in a huge number of variant processes. Process experts from each domain, like quality and testing, assembly departments and manufacturing units, had to spend significant time in analyzing influencing factors for malfunctioning and deciding on best action to prevent end of line test failures.
Therefore, an AI project was started with the objective to get reliable and fast results on root cause discovery. Speed is important because production runs 24 hours / 7 days a week. The sooner the real reasons for mal-functions are discovered, the sooner activities can be implemented to avoid bad quality. This saves a lot of time and reduces significant waste. The target is to reduce waste in certain manufacturing domains by 20%. The key success factor is the fast detection mechanism within the production chain delivered by AI.
Complex root-cause findings can be reduced from several days to hours.
ZF's intention with the digitalization approach is to deliver fast information to the people who are responsible for decision processes to keep a plant in an optimal output with high quality products. A self-learning AI solution Predictive Intelligence from IS Predict was used to analyze complex data masses from production, assembly and quality and to find reliable data patterns, giving transparency on disturbing factors / factor combinations. For training algorithms, end to end tracing data was used, made available in a data lake.


Quantifying the impact of Customer Experience Improvements on Churn vial Causal Modeling

Uhrzeit: 15:30 - 16:15
Vortrag: Di 3.3 2)


Quantifying the impact of customer experience (CX) improvements on the financials is crucial for prioritizing and justifying investments. In telecommunication as well as other subscription-based industries, churn is one of or the most important financial aspects to take into account. The presented approach shows how the churn impact of CX improvements - measured via Net Promoter Score (NPS) - can be estimated based on structural causal models. It makes use of algorithms for causal discovery and counterfactual simulation.

Target Audience: Data Scientist, Decision Maker
Prerequisites: Basic understanding of statistical modeling
Level: Advanced