Charles Douse
Senior Data Scientist @Marks and Spencer
Signup · Get unlimited contacts
WORK HISTORY
Senior Data Scientist @Marks and Spencer
London, GB
Collaborating with supply chain teams to develop a simulator that replicates current allocation and replenishment processes. This tool will enable users to perform scenario planning by adjusting parameters such as safety stock and cover duration to see how changes affect recommended shipments, stock levels, lost sales, and overall sales (using unconstrained sales calculations). Additional uses include diagnostic analysis, value measurement (by assessing the impact of improved forecast accuracy), and optimisation of supply chain parameters within the existing network. Charles is also exploring a forecasting solution for new products that lack historical sales data. Opportunities identified to automate and improve forecasting accuracy of online sales to introduce benefits such as better labour planning in distribution centres and reduced customer delivery times. Charles built a forecasting tool using Prophet that improved MAPE in Q3 2022 (peak period) versus previous manual forecasting by more than 3%. Solution pipeline developed and deployed in Azure Data Factory, with daily automated reporting.
EDUCATION
Dame Alice Owen's School
A-Levels, Mathematics, Further Mathematics, Physics, Economics
UCL
Master of Science (MSc), Mathematics
SKILLS
ABOUT CHARLES DOUSE
Charles is a Senior Data Scientist at Marks & Spencer. He is building a supply chain simulator that replicates how products are allocated and replenished within stores at M&S. Charles previously built a forecasting tool for online orders to optimise labour planning at distribution centres. At Cognizant, Charles implemented gradient boosting algorithms to predict faults and model degradation. He has a track record of developing and deploying large-scale algorithms that have significantly impacted business revenues and user experience. Charles also has strong experience building end-to-end machine learning solutions using Big Data technologies (Spark and Hive).Charles was recently an Analytics Manager, where he conducted statistical hypothesis tests for Google to determine the success of marketing campaigns by measuring performance against key success factors. He built a linear regression model to predict incremental sales from media activity on a major online marketplace. The results were used by the client to inform their media spend in several markets.At Capgemini, Charles managed various Agile projects at a large government department and built an automated finance model whilst working in one of the world\'s largest IT transformation programmes at a multinational banking and financial services company. He worked at the biggest retail bank in the UK and managed a team of 5-10 consultants to support the bank with the end to end delivery of data requests; including support with data delivery planning, extraction, analysis, MI, and development of automated financial models & calculators. Charles has technical abilities covering a wide range of analytical software tools, including Oracle SQL Developer, BigQuery and Teradata for data extraction, Microsoft Excel and SAS for data analysis, programming languages such as Python, R and VBA, data visualisation tools such as Microsoft Power BI and Tableau, and machine learning techniques (linear regression, logistic regression, random forest, neural networks, k-means clustering).He previously worked as a Data Analyst within the Supply Chain and Logistics teams at Sainsbury\'s. Charles is competent in building and developing complex business driven models.Charles is a MSci Mathematics graduate from University College London with First Class Honours.He travelled extensively for 10 months after graduating from university. This included working in Sydney, Australia for almost 5 months in different job positions and a very unique work environment.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.