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

SOFTWARE MEETS BUSINESS:
Die Konferenz für Software-Architekturen
30. Januar - 03. Februar 2017

Sessionsdetails

Vortrag: Mi 4.3
Datum: Mi, 28.10.2020
Uhrzeit: 15:10 - 16:40
cart

Sales Forecasting & Stock Optimization with Machine Learning and Self Service BI at Salus Group

Uhrzeit: 15:10 - 15:55
Vortrag: Mi 4.3 1)

 

Salus is specialised to provide complete distribution, promotion and services for medicinal product or device.The scope was sales forecasting lead to better understanding of patterns of item level. Those data is used to optimize frequency of ordering and ordered quantities.The challenge was the huge number of items and ordering frequencies leading to big complexity. Those complexity leading to frequent stock-outs or to high dead stock. Our developed solution is in charge for creating ordering scheduler handling order times and quantities.

Target Audience: Decision Makers, Data Engineers, Data Scientists and Project Leaders in AI- and BI-projects
Prerequisites: Interested in machine learning-projects and self service BI, understanding in distribtion requirements, logistic processes or retail
Level: Basic

Extended Abstract
The Salus Group is specialised to provide complete distribution, promotion and active sales services as well as value added services necessary for the medicinal product or medical device to be placed on the market it represents a vital link in the supply of medicinal products, dietary supplements, medical devices and high-quality and innovative services, which provide health and well-being of people.

The project scope was the sales forecasting lead to better understanding of patterns on item level. Those data is used to optimize frequency of ordering and ordered quantities.

The Challenges:

  • the huge number of items and ordering frequencies leading to big complexity.
  • the manual analysis and manual ordering can leave outliers and patterns undetected
  • those complexity leading to frequent stock-outs (lost sales) or to high dead stock (cash flow impact)

The Solution:

  • Develop algorithm trained on history data combining external data sources as weather in order to forecast sales
  • Create algorithm that is in charge for creating ordering scheduler handling order times and quantities
  • Maintain whole system with stock monitor which can detect outliers in sales and trigger near-real-time ordering

The Result:

  • Number of stock-outs minimized
  • High savings in cash flow in first year thru lower dead stock
  • Automatic integration with ERP system saving employees time
  • Clear interpretability of given results thru self-service analytics in Qlik Sense

 

OmniChannel-Transparenz: So hilft der Commerce Reporting Standard beim Auswerten vom Filialgeschäft

Uhrzeit: 15:55 - 16:40
Vortrag: Mi 4.3 2)

 

OmniChannel ist das Zauberwort. Online-Prozesse datengetrieben zu steuern, ist heutzutage Best Practice. Die Schwierigkeit liegt darin, diese mit Offline-Daten zu verknüpfen. Der Vortrag erläutert, wie der CRS dieser Problematik Abhilfe schafft.

Zielpublikum: BI-Projektleiter, Business & Data Analysts, Consultants
Voraussetzungen: Basiswissen
Schwierigkeitsgrad: Anfänger