Führen Sie Ihr Geschäft mit datengesteuerten Erkenntnissen

2100 man-hours, 1+ years

U-net, OpenCV, Postgres, CouchDB

Investigate whether it is technically and commercially feasible through machine learning to obtain information about location and composition of soccer fields.

We built an application to select areas of interest, obtain a list of football fields, and view the results using a map application. A distinction is made between the fields covered with natural grass and artificial turf.

–

SCRUM, Kanban, SAFe elements

How to roll-out the solution throughout the organization in 40+ countries while constantly developing new features?

We combined the main teams running SCRUM with supporting teams in Kanban mode. At the same time using SAFe PI planning methodology for planning & roadmaps across the teams, including the business stakeholders.

• Developing the DWH while constantly extending over new units in the organization and extending the user base.
• Running roadmap for upcoming 12 months, adjustable for future roll-out needs.
• Sprint plans for upcoming quarter, dependency mapping.
• Combining the best practices of BI and software development provides real flexibility.

• Rolled out in 120+ units worldwide with 420+ master data managers and still constantly expanding
• Running 5 teams in parallel combining SCRUM, Kanban and SAFe elements
• Business stakeholders have clear visibility and predictability of the investment value

Snowflake, Airflow, dbt, Power BI​

Our client began implementation of a new Snowflake DWH system three years for advanced analytic scenarios and to support business decision making and reporting. ​

Soon the amount of manual work due to the reporting quality had expanded significantly. ​

Because the DWH did not fulfill the client’s business needs, and ascertaining what causes the issues was starting to cause considerable overhead, they decided to seek help. ​

• Architecture that was flexible, where changes could be administered from a single place.
• Helped the client build a foundation and processes for a dimensional data model for their DWH. ​ ​
• Implemented new technologies such as Airflow and dbt. This now serves as the client’s roadmap and helps them keep pace with business growth.​ ​

• Simplified business reporting logic and efficient data retrieval.​

• Clarity, understanding, and reusability. Flexibility to changes in business. ​

• Ease of validating data integrity.​

Oracle PL/SQL, Data Vault modelling​

A central database to track operator productivity in hundreds of warehouses across the globe. ​

The client needs to track the productivity, in order to identify bottlenecks, adjust the workload and also forecast for better future performance.​

• A data warehouse based on data vault principles to allow dynamic management of business rules without compromising data integrity. ​ ​
• A reporting source for physical warehouse operator productivity calculations.​ ​
• Due to multiple regulations from a variety of countries, the solution has to be adaptable.​

• International, country-specific operations reports are handled within one single system.​​

• Reliable operations data about warehouse sites to measure and forecast productivity.​ ​

• 80 operations implemented the solution during the first year of development.​

Data Lake, DWH, Data integration toolsets, Metadata management, Change data capture (CDC), Scheduling and workflow toolsets​

Our client was challenged with choosing the right data platform stack to serve them for years to come. ​

They were faced with numerous modern cloud technologies on the market, a deadline set in stone, and internal resources already engaged for the next nine months.​

Over 300 scenarios were defined and run on competing platforms. Technology combinations were evaluated for suitability in the bank’s business context. ​

Proekspert’s assistance enabled the bank to choose a data platform stack partner based on total cost of ownership and the partner’s compatibility with their culture. ​

• We helped the client to make an educated decision for their future DWH platform.

• Deep dive into the truth behind promotional materials from different vendors.​

Azure, Power BI, MongoDB​

Huvepharma is providing pharmaceuticals for farm animal facilities. ​

In order to pinpoint problems near real time, they needed to have aggregated information about the health of client facilities, they needed to provide an integrated solution in their end-user app.​

• Clients enter the assessment data in their facilities.​
• Data is immediately aggregated and displayed in embedded reports for quick decision making.​

• Huvepharma customers can make quick decisions based on their own data.​

• Ability to provide custom reports for clients on the go.​

• Ability to see the patterns in client facilities to identify health deviations.​

Several image recognition technologies and methods​.

In cooperation with ESA: read more here →

We needed to cover 2 cases, based on detected roof geometry from satellite images:​

• Calculate the potential cost for the solar roof​
• Forecast the productivity of the planned solar roof​

By combining these, sales team and the potential customer can make an educated investment decision.​

Using analysis of Earth Observation imagery, calculate the areas and angles of roofs at given address.​

Combining Earth Observation data with other geoinformatic data sources can further improve the results.​

• Our client can now get a baseline ROI indication for potential customers quickly, increasing competitive advantage.​

• Adding new geographical areas is simple and only minimal additional training for the model is required.​

• Earth Observation imagery can be combined with other geoinformatic data sources data to enhance the accuracy.​