Resource Planning
Minor Data Driven Business Lab
Client company:Cohesive
Mariela Gocheva
Bart Mioch
Urban Jasurek
Monique Chandra
Braulio Varges
Aimar Nebreda
Project description
How can we provide insights into how Cohesive can manage their resources best given the data about the kinds of projects and the talent pool they have?
Context
Cohesive is a business consultancy firm facing challenges in managing and allocating its workforce of over 900 consultants. The company struggles with assigning the right consultants to the right projects, particularly when priorities shift, leading to inefficiencies and project delays. Additionally, the lack of forecasting insights makes it difficult to predict future resource needs. This project aims to analyse Cohesive’s current resource management process and propose data-driven solutions to improve consultant allocation and enhance forecasting.
Results
Through comprehensive data analysis and the application of various data-cleaning techniques, we successfully organized the datasets and prepared them for deeper exploration. By linking different components of the data — such as skills data to corresponding resources — we created an overview of currently available resources as well as those projected for future months. This information was made interactive and accessible through a custom-built PowerBI dashboard, allowing users to filter resources by time period, location, and role.
Additionally, we explored forecasting options by training basic models, including linear regression, the Prophet model, and ARIMA, to predict future trends and resource demand.
About the project group
Our team brings together a diverse range of backgrounds, with expertise in software engineering, machine learning, business logistics, and even musical engineering. This blend of skills created a dynamic and creative environment as we collaborated on the project over the past five months. Throughout the process, we adopted the Agile Scrum methodology, which helped us set clear goals and achieve them through iterative cycles of analysis, implementation, review, and refinement.