Abstract

Supply chains represent the next frontier in Big Data. In an increasingly competitive environment, the need for supply chain optimization and real-time supply chain management becomes evident. For proper design and evaluation of forecasts, orders, and other supply chain parameters, comprehensive and timely information is essential. Significant operational improvements can be made by ensuring the proper functioning of inventory flows across the supply chain. Our overarching goal is to address challenges in real-time inventory management in Supply Chain Networks using Data Engineering. A wholesale distribution network with a warehouse and retailers is considered as a case study. We work with data-rich situations, where there is a wealth of information available in the form of incoming orders at the retailers.

In summary, the goal is to estimate demand parameters (lead-time, post-lead time distribution), optimize order schedules (policies) such that order quantities are minimized holding $g$-costs, and develop infrastructure and deployment solutions for proper implementation of the Case Study. The emphasis lies on parameter estimates. Each retailer places an order according to a policy which is the main focus of this work. Extensive focus is put on the database aspect, which estimates nonlinear delay and time-lag of the up-to-order retailer. A scripting language processes up-to-order information from any retailer and ensures that the coded and prefiltered order data can be easily integrated to different Enterprise Resource Planning Systems or utilized in any commercial MRP-Inventory Management Framework.

Overall, understanding and manipulating the order data can readily lead to significant reductions in behavioral costs associated with orders that are not properly scheduled. Moreover, improved estimates can serve as input parameters in other estimators or optimizers that involve larger groups of retailers and their orders.

Keywords

  • Real-Time Data Processing
  • Inventory Optimization
  • Big Data Analytics
  • Wholesale Distribution
  • Supply Chain Visibility
  • Stream Processing
  • Demand Forecasting
  • Data Lake Architecture
  • Apache Kafka
  • Predictive Analytics
  • IoT Integration
  • Distributed Systems
  • Data Pipeline Automation
  • Machine Learning Models
  • Operational Efficiency.

References

  1. 1. Serón, M., Martín, Á., & Vélez, G. (2019). Life cycle management of automotive data functions in MEC infrastructures. arXiv. https://arxiv.org/abs/2303.05960
  2. 2. Pillmann, J., Wietfeld, C., Zarcula, A., Raugust, T., & Calvo Alonso, D. (2018). Novel Common Vehicle Information Model (CVIM) for Future Automotive Vehicle Big Data Marketplaces. arXiv. https://arxiv.org/abs/1802.09353
  3. 3. Berezovsky, A., El-khoury, J., Kacimi, O., & Loiret, F. (2018). Improving lifecycle query in integrated toolchains using linked data and MQTT-based data warehousing. arXiv. https://arxiv.org/abs/1803.03525
  4. 4. Capgemini. (2018). Premium OEM leverages IoT-based data in the Cloud to track vehicles in distribution. Capgemini. https://www.capgemini.com/news/client-stories/premium-oem-leverages-iot-based-data-in-the-cloud-to-track-vehicles-in-distribution/
  5. 5. Shiklo, B. (2017). IoT in the Automotive Industry: Concept, Benefits, and Use Cases. ScienceSoft. https://www.scnsoft.com/blog/iot-in-automotive-industry