Publication
Supply Chain Optimization Using Python and Mathematical Modeling
https://www.toptal.com/developers/mathematics-expert/supply-chain-optimization-product-blending
Michael is currently unavailable
Michael Hopf
Verified Expert in Engineering
Artificial Intelligence Developer
Düsseldorf, Germany
Toptal member since May 3, 2022
Bio
Michael is a specialized consultant working on DS projects, especially in the logistics and manufacturing sectors. He's worked on various supply chain optimization projects improving logistics for heavy industry clients like rail, vessels, and mining. He's worked with advanced AI projects involving object detection and NLP. Michael has a PhD in mathematical optimization and was a management consultant in the analytics practice of McKinsey & Company, currently Quantum Black, for four years.
Portfolio
FOXI abrechnung
Data Science, Artificial Intelligence (AI)
Haymaker Content LLC
Python, Adobe ExtendScript Toolkit, OpenAI GPT-4 API
Freelancer
Python, Excel 365, SAP, SAP Business Warehouse (BW), TM1, SQL...
Experience
- Optimization - 10 years
- Deep Learning - 8 years
- Machine Learning - 8 years
- Data Science - 8 years
- Python - 8 years
- Generative Pre-trained Transformers (GPT) - 5 years
- Computer Vision - 5 years
- Natural Language Processing (NLP) - 5 years
Preferred Environment
Windows, PyCharm, Microsoft PowerPoint, Excel VBA, Excel 365, Jupyter Notebook, Python, GitHub, Amazon Web Services (AWS), Azure
The most amazing...
...solution I've developed is a train scheduling software for a large coal mining company in Canada with complex product mixing constraints.
Work Experience
Senior Developer
2023 - 2025
FOXI abrechnung
- Collaborated on creating software that helps automate the documentation process for dentists in Germany.
- Worked on the developed AI that has to understand and apply complex legal requirements for documentation.
- Had to scale to the required usage as a SaaS software.
Technologies: Data Science, Artificial Intelligence (AI)
Senior Developer
2023 - 2023
Haymaker Content LLC
- Worked on building an extension for Adobe Premier for an AI product. More than half of the production cost for reality TV shows occurs in post-production, especially in creating cuts from the recorded video clips.
- Developed an Adobe Premiere plugin that incorporated the ChatGPT API to ingest a description for a desired scene, then created a rough cut by linking this summary to the transcript and the connected video clips.
- Worked on an end-to-end solution and AI workflow for reality TV rough cuts.
Technologies: Python, Adobe ExtendScript Toolkit, OpenAI GPT-4 API
Data Science Consultant
2022 - 2022
Freelancer
- Restructured processes to estimate raw material prices and create forecasts for the third-largest printing color producer in the world.
- Developed a new analytically driven process to select a subset of raw materials for manual price forecast by experts.
- Created a method for landed cost simulation to incorporate transport cost and duty cost into business calculations.
- Improved communications between different departments of the company by redefining ways of interactions.
Technologies: Python, Excel 365, SAP, SAP Business Warehouse (BW), TM1, SQL, Microsoft Power BI, Scikit-learn, NumPy, Pandas, Forecasting, Time Series Analysis, XGBoost, Power Query, Snowflake, Linear Optimization, Algorithms, Supply Chain Management (SCM), Supply Chain Optimization, Leadership, Vehicle Routing, AI Agents, Vector Databases, Retrieval-augmented Generation (RAG)
Lead Data Scientist
2021 - 2022
Freelancer
- Worked for a US startup that develops software for analyzing remote meeting calls like Zoom with AI methods.
- Developed an NLP algorithm to detect questions of different types in conversations and their corresponding answers.
- Built an NLP algorithm to track custom-defined topics in sales conversations.
- Developed an NLP algorithm to segment transcribed conversations into semantically connecting sections.
Technologies: Python, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), AWS Lambda, Docker, Deep Learning, Machine Learning, Scikit-learn, NumPy, Pandas, CTO, Text Analytics, BERT, Amazon Web Services (AWS), Databases, Data Engineering, Startup Consulting, Linear Optimization, Algorithms, OpenAI GPT-4 API, OpenAI GPT-3 API, Retrieval-augmented Generation (RAG)
Specialist in Data Science
2018 - 2022
McKinsey & Company
- Created a network strategy for a global logistics player leading a small team to develop a strategy to combine postal and parcel networks in a novel way.
- Optimized underground rail network system for a platinum mine in South Africa and increased trammed material by 10%.
- Led a team of three data scientists in an analytics engagement to maximize the gas throughput of a gas platform in the North Sea, which led to a 2% increase in production.
- Optimized domestic US shipping for global oil and gas clients for the US Gulf barges market.
- Optimized throughput for an agriculture client at the biggest plant in Europe for oil extraction of soy and canola beans.
- Optimized a rail network for a large Canadian mining company being the lead data scientist of two teams with more than 15 data scientists and developers involved.
- Developed algorithms for the next best action recommendations for digital marketing strategies for a global pharmaceutical client.
- Developed AI image recognition solutions for automated pothole detection for a large construction company.
Technologies: Python, Amazon Web Services (AWS), Excel 365, GitHub, Microsoft PowerPoint, PyCharm, Jupyter Notebook, Azure, Computer Vision, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Machine Learning, Data Analysis, Data Analytics, Artificial Intelligence (AI), TensorFlow, Amazon S3 (AWS S3), Amazon EC2, Object Detection, Image Processing, Scrum, Scikit-learn, NumPy, Pandas, XGBoost, Microsoft Excel, Power Pivot, Data Visualization, Clustering, Neural Networks, Sensor Data, JSON, ETL, Time Series, Tableau, Convolutional Neural Networks (CNNs), Business Analysis, Statistical Modeling, Dashboards, Predictive Modeling, Statistics, REST APIs, Technology Consulting, Advisory, R, Azure Cognitive Services, OpenCV, Computer Vision Algorithms, Linear Optimization, Algorithms, Google OR-Tools, Supply Chain Management (SCM), Supply Chain Optimization, Leadership, Vehicle Routing
Research Assistant
2014 - 2017
Universitay of Kaiserslautern
- Worked as a research assistant in the optimization research group.
- Gave lectures and tutored in nonlinear optimization, and supervised master's theses.
- Founded the optimization elite group to guide exceptional students towards a PhD.
Technologies: Optimization, Nonlinear Optimization, Scheduling, Packaging, Research
Intern
2014 - 2014
DZ Privatbank
- Worked in the treasury department as a foreign exchange trader.
- Performed over-the-counter (OTC) trading and validation of foreign exchange (FX) swaps.
- Developed and programmed VBA macros for the automated processing of transactions.
Technologies: Treasury, Trading, Forex Trading, Forex Trading Room, Forex Analysis
Intern
2013 - 2013
Krankenhaus der Barmherzigen Brüder
- Developed a tool for the analysis of clinical pathways in Visual Basic.
- Analyzed patient waiting times and equipment idle times.
- Contributed to staff planning and quality management.
Technologies: Visual Basic, Excel VBA, Excel 365
Experience
Optimization of Coal Supply Chain
We optimized a large Canadian mining company's supply chain.
From mine to port, supporting a $1 billion investment for a new port. I was the lead data scientist of two teams with more than 15 data scientists and developers involved in managing the development of two planning tools that used optimization algorithms to determine the best train schedules, port schedules, and coal product blending.
In particular, product blending (mixing) became apparent as a significant impact lever with considerable opportunities to optimize mixing rules on components or essay qualities.
The final solution was a dashboard where the rail planners received recommendations on which schedule achieves the greatest throughput and why.
From mine to port, supporting a $1 billion investment for a new port. I was the lead data scientist of two teams with more than 15 data scientists and developers involved in managing the development of two planning tools that used optimization algorithms to determine the best train schedules, port schedules, and coal product blending.
In particular, product blending (mixing) became apparent as a significant impact lever with considerable opportunities to optimize mixing rules on components or essay qualities.
The final solution was a dashboard where the rail planners received recommendations on which schedule achieves the greatest throughput and why.
Optimization of Underground Rail
We diagnosed and optimized an underground supply chain from blasting faces to hoists of a large South African platinum mining company.
We improved battery changing strategies and reduced loco congestion to increase the trammed material by 10%. I was the lead data scientist for the development of the planning tool and managed the implementation of change management for the loco crews underground.
We improved battery changing strategies and reduced loco congestion to increase the trammed material by 10%. I was the lead data scientist for the development of the planning tool and managed the implementation of change management for the loco crews underground.
Optimization of Postal Network
We created a new network strategy for a global logistics player to deal with the exploding eCommerce demand during the COVID-19 crisis.
I led a small team, and we developed a strategy to combine and utilize both post and parcel networks in a novel way. We tested the approach with an analytics network model and convinced the CEO to implement the new strategies.
I led a small team, and we developed a strategy to combine and utilize both post and parcel networks in a novel way. We tested the approach with an analytics network model and convinced the CEO to implement the new strategies.
Education
2014 - 2017
PhD in Mathematics
University of Kaiserslautern - Kaiserslautern, Germany
2012 - 2013
Master's Degree in Mathematics and Computer Science
University of Kaiserslautern - Kaiserslautern, Germany
2011 - 2011
Master's Degree in Mathematics and Computer Science
University of Singapore - Singapore, Singapore
Skills
Libraries/APIs
Scikit-learn, NumPy, Pandas, XGBoost, TensorFlow, OpenCV, REST APIs, Azure Cognitive Services
Tools
Microsoft PowerPoint, GitHub, Gurobi, Microsoft Excel, Google OR-Tools, PyCharm, Microsoft Power BI, TM1, MATLAB, Power Pivot, Power Query, Tableau, Adobe ExtendScript Toolkit
Languages
Excel VBA, Python, Visual Basic, SQL, R, Snowflake
Platforms
Jupyter Notebook, Windows, Azure, AWS Lambda, Docker, Amazon EC2, Amazon Web Services (AWS)
Paradigms
Scrum, ETL, Linear Programming
Storage
JSON, Amazon S3 (AWS S3), Databases
Other
Excel 365, Optimization, Data Science, Integer Programming, Natural Language Processing (NLP), Nonlinear Optimization, Packaging, Treasury, Logistics, Last Mile Logistics, Mathematics, Applied Mathematics, Networks, Machine Learning, Deep Learning, Computer Vision, Artificial Intelligence (AI), Data Analysis, Data Analytics, Neural Networks, Text Analytics, Convolutional Neural Networks (CNNs), Predictive Modeling, Technology Consulting, Advisory, Startup Consulting, Linear Optimization, Algorithms, Supply Chain Management (SCM), Supply Chain Optimization, Generative Pre-trained Transformers (GPT), Vehicle Routing, Analytics, Forex Trading, Graphs, Object Detection, Image Processing, Forecasting, Time Series Analysis, BERT, Time Series, Business Analysis, Statistical Modeling, Dashboards, Statistics, Computer Vision Algorithms, Leadership, OpenAI GPT-4 API, OpenAI GPT-3 API, AI Agents, Retrieval-augmented Generation (RAG), Recommendation Systems, SAP, SAP Business Warehouse (BW), Scheduling, Research, Trading, Forex Trading Room, Forex Analysis, Philosophy, Data Visualization, Clustering, Sensor Data, CTO, Data Engineering, Vector Databases
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring