Mondelēz International needs to react to changing business conditions with excellent data analysis, define requirements, perform analysis, and identify patterns in large amounts of data to improve demand forecasting accuracy and bias.
Requirements
- Design, build, implement and/or maintain Python/PySpark algorithms to execute predictive analysis which help to improve planning processes.
- Analyze code, define, and implement strategies to have more accurate and efficient results
- Working in cloud computing environments with tools like databricks and help MDLZ to have better practices on it.
- Determine, create and maintain the best Statistical models be to be used, by considering SKU demand behavior using segmentation strategy, to generate high quality demand statistical forecast with low forecast error and bias
- Refine forecasting models, by reviewing forecast performance and incorporating feedback from the Demand Planner, to improve forecast error and bias metrics
- Analyze the model performance every month / week where MAPE (Main Absolute Percentage Error) is deteriorating and post process the output and if required finetune the output
- Experience in any Data Science language like Python (preferred), PySpark (preferred), R, Julia.
Responsibilities
- Lead statistical forecasting process to provide business a statistical base forecast for its demand planning process.
- Lead some continuous improvement projects to get a more efficient, robust, and accurate models to predict.
- Analyze business requirements as a guide for data modeling and apply data analysis, design, modeling, and quality assurance techniques, based on a detailed understanding of business processes, to establish, modify or maintain data structures and associated components (entity descriptions, relationship descriptions, attribute definitions
- Manage the iteration, review and maintenance of data requirements and data models and you will assist in creating the sematic layer of data
- Choose a suitable data modeling approach for each project, such as by assessing the suitability of existing data models and build data models with the flexibility to change when business requirements change
- Reconcile multiple logical source models into a single, logically consistent model and you will ensure that the proposed model follows data architecture guidelines and best practices
- Design, build, implement and/or maintain Python/PySpark algorithms to execute predictive analysis which help to improve planning processes.
Other
- Extensive experience in data modeling in a large complex business with multiple systems
- Ability to simplify complex problems and communicate them to a broad audience
- Aligning and coordinating activities with different areas and stakeholder to guarantee a good execution of projects and processes.
- 10+ years of experience in handling multiple projects leading in the domain of supply chain and forecasting with multiple team members as part of org structure
- Fluent English is a must
