A company in the healthcare sector with over 1,500 pharmacies wanted to better anticipate its sales in order to optimize its stocks, limit product shortages and adjust its orders to actual needs.
💡 What I’ve done
Analysis of sales trends over several years to detect seasonal variations
Segmentation of sales outlets into 4 profiles (urban, rural, high traffic, low traffic)
Setting up two forecasting models :
✅ A simple method: 15-day rolling average
🔍 An advanced method: modeling time series with Prophet
Comparison of model performance to select the most reliable approach for each profile
📈 The results
🔹 Better understanding of store profiles 🔹 More precise forecasts by type 🔹 A solid basis for better inventory management and logistics adaptation
In short: finer decisions, fewer disruptions and optimized management 📦
A supplier to company restaurants and canteens is looking to expand its fresh fruit and vegetable business. But past logistical problems have led to fears of stock-outs and delivery delays, particularly during peak demand periods.
💡 What I’ve done
Identification of the top 5 most ordered products: tomatoes, apples, bananas, apricots, kiwis
Analysis of sales over the last 4 years, by product reference
Study of seasonal variations (summer/winter) for each product
Cross-cycle analysis to identify products with high co-sales or to stock together
Recommendations for reorganizing the warehouse according to cycles and optimizing flows
📈 The results
🔹 Better visibility on future volumes 🔹 A clear strategy for seasonal warehouse storage 🔹 Concrete ways to reduce delays and secure customer satisfaction
A cereal bar manufacturer produces multi-variety boxes. When only one type of bar is lost (due to quality or machine failure), the whole box is unusable. The factory wanted to measure this imbalance precisely, so as to be able to make concrete decisions.
💡 What I’ve done
Industrial data cleansing from two production lines (duplicates, inconsistent names, raw structure)
Clean database with SQL structuring
Advanced business queries to answer operational questions
Excel-based data modification tool, ready for use by on-site teams
Interactive dashboard built on Power BI with :
Imbalance percentage
ATS (Adherence To Schedule)
Number of bars produced
Number of boxes produced and disposed of
📈 The results
🔹 A better understanding of losses on multi-variety boxes 🔹 A visual tool to control production and exclude anomalies 🔹 Concrete ways to adjust industrial processes
🧰 What about the technical side ?
Advanced SQL for structuring and queries (aggregations, complex conditions)
Power Query + Excel for cleaning + manual exclusion
Power BI to create the final dashboard
Work with real, imperfect and complex production data
🤝 Let’s talk about it !
Do you work with data from industry or the supply chain?
I can help you highlight your inefficiencies, gain visibility and manage your performance with concrete tools that your teams can use.
An industrial company operating on an international scale was facing high logistics costs and unstable delivery times between its factories in Africa and its end customers.
The objective: to rethink the organization of logistics flows while maintaining a high level of customer service and preserving sales.
💡 What I’ve done
Analysis of current logistics flows: plant → port → customer assignments
Assessment of carrier performance: compliance with contractual lead times (max. 72 hours)
Complete mapping of logistics roads and friction points
Calculation of total daily costs (production + transport)
Optimize distribution flows with Power Query to reduce lead times and streamline flows
Elaboration of alternative scenarios and concrete recommendations for Supply Chain management
📈 The results
🔹 Clear view of logistics organization and areas for improvement 🔹 Cost savings estimated at 13% on optimized flows 🔹 Scenario to deliver all orders in 11 days 🔹 Realistic recommendations on factory upgrades and carrier contract negotiations
🧰 What about the technical side ?
Exploratory and bivariate analysis in Python
Data processing and transformation with Excel + Power Query
Optimization methodology inspired by graph theory
Logistics mapping and performance visualization in a clear presentation (PowerPoint)
🤝 Let’s talk about it !
Do you manage an international or multi-site supply chain? I can help you with :
Map your logistics flows
Evaluate your suppliers
Optimize your supply chain
Visualize your performance using simple decision-making tools
A fast-growing chain of DIY stores, was experiencing frequent stock-outs, especially during periods of high demand.
Procurement teams needed a simple, customizable and automated tool to place the right orders at the right time, and gain operational reliability.
💡 What I’ve done
Development of an Excel-based DRP (Distribution Resource Planning) tool:
Calculating projected inventory over several weeks
Visualization of stock coverage
Follow-up of confirmed and planned orders
DRP V2 integrating:
Dynamic safety stock (adjustable in days)
MOQ (Minimum Order Quantity) per item
Parameters that can be adjusted directly by suppliers
Add a clear, visual histogram to monitor stock variations
Design a training manual for the tool
📈 The results
🔹 An automated, educational tool, 100% Excel, tailored to the business 🔹 Fewer disruptions, better anticipation, greater comfort for the team 🔹 Structured, easily modifiable order logic 🔹 Better appropriation of management rules: MOQ, safety, transfer
A multi-product online sales company (tech, home, toys…) was faced with an increasing number of customer returns, with no real capacity for analysis. The data was stored in a single, poorly-readable file, making it impossible to :
Precise tracking of returns
Identification of recurring causes
And implementation of reliable indicators for action
💡 What I’ve done
Comprehensive data dictionary for structuring information (customers, orders, products, refunds, returns, etc.)
Relational schema to model the database
Building a database in SQLite
Drafting and execution of SQL queries to meet business needs (identification of reasons for returns, value of refunds, etc.).
Clear, accessible presentation with :
5 major reimbursement issues + their solutions
5 realistic ideas for recycling returned products (ecology, circular economy, brand image…)
📈 The results
🔹 Creation of a structured basis for customer feedback analysis 🔹 Highlight the main causes of refunds (e.g. damaged product, preparation error, unclaimed package, etc.). 🔹 Actionable recommendations to reduce returns and enhance the value of damaged products 🔹 Clear, comprehensible support for all departments (logistics, customer service, management)
A French textile company committed to local, sustainable production wanted to :
Cleaning up its data in compliance with the GDPR
Extract useful HR indicators without exposing sensitive information
Analyze the performance of logistics flows (budgets, CO2, quality)
Compare its current transport with a more ecological and economical alternative
💡 What I’ve done
GDPR cleansing of TMS data (anonymization of social security numbers)
Extraction of HR information (gender & year of birth) for anonymized statistics
Creating an HR dashboard :
M/W distribution in pie chart
Age brackets by histogram
Construction of a 9-month cumulative logistics dashboard :
Tracking of transported load, distances, costs and CO2e emissions
Year N / N-1 comparison
Comparative budget study between current carrier and new carrier proposal :
Integration of tariff grids + diesel surcharges
Analysis of transport orders from July to September 2022
Environmental impact assessment of vehicles used :
Comparative CO2e balance
Recommendation for transition to lower-emission vehicles
📈 The results
🔹 HR data that can be used without compromising privacy 🔹 Clear dashboard for monitoring transport performance 🔹 New carrier solution identified as more economical (-4.23%)
🔹 Recommendation validated for greener transport 🔹 Support ready for presentation to CSR Committee and Supply Chain Management
A company delivering household goods was faced with a constant flow of items back and forth between stock and packaging areas.
Certain high-demand products were also frequently returned, creating a significant loss of time for on-site teams.
The aim was to identify fast-moving items, so as to consider investing in an intelligent sorter capable of streamlining the flow of goods.
💡 What I’ve done
ABC analysis of shipments (outgoing) and returns to identify the most affected items
Visualize Pareto curves to analyze distribution
Cross-referencing of shipment/return results with application of a rotation score
A/B/C weighting → Calculation of a rotation index
Final classification of products with the highest sorting potential
Creation of an interactive Power BI dashboard, with :
Global time filtering
Analysis by returns, departures or turnover rate
View by product category or individual item
📈 The results
🔹 Identification of critical rotation products, justifying pre-storage in the sorter 🔹 Better visibility for operators: focus on 5 key products 🔹 A scalable tool for tracking volumes and fine-tuning logistics strategies 🔹 Projected time savings through intelligent product flow management
🧰 What about the technical side ?
Excel for initial ABC analysis
Power BI for final dashboard creation
Rigorous methodology based on Pareto’s law
🤝 Let’s talk about it !
Do you manage a warehouse, a supply chain, or a large volume of product returns? I can help you with :
Identify high-turnover products
Set up a logistics tracking dashboard
Help your teams save time and optimize operational flows
In a fast-food chain specializing in local fresh produce, the main warehouse experienced a sharp increase in stock (+89% in value) in 9 months, with no apparent cause.
Management wanted to understand this phenomenon in order to avoid bottlenecks, better plan supplies, and ensure the safety of their teams.
💡 What I’ve done
Detailed analysis of 6 key indicators, based on 4 datalake sources (orders, receipts, inventory, shipments) :
Quantity in stock: +67% between May 2022 and January 2023
Average purchase price : -9.44% → not the cause of the increase in value
Number of references stocked : +7,95 %
Quantities received vs. shipped: +20% vs. +27% → relative imbalance
Number of orders placed : -9.24% (but overall upward trend)
Supplier service rate: +5.71 points → significant improvement
Build a dynamic Excel file with :
Clear charts for each indicator
Calculation of variation between 1st and last month
Filtering by product family
Identification of the 3 families with the most pronounced stock increases :
Bakery
Vegetables mix
Dairy
Preparation of a clear, educational visual aid (PowerPoint) for distribution to non-technical teams
📈 The results
🔹 Precise visualization of stock evolution factors 🔹 Putting the purchase price out of the equation (observed drop) 🔹 Identification of 3 families with high stock drift 🔹 Targeted recommendations by family, including :
