Large professional services companies employ thousands of professionals making labor the industry’s highest expense. Examples of industries where professional service firms are commonly used include: Information technology outsourcing, Business process outsourcing, Accounting, Legal, Advertising, Engineering, and other specialized services.

Companies are using Machine Learning to automatically compute the matching score between resources and jobs, but the actual assignment is done manually which leads to poor demand fulfillment and low labor resource utilization, high project delivery costs and low customer satisfaction.

The integration of Machine Learning (ML) and Mixed Integer Programming (MIP) is essential to solve Resource Matching Optimization (RMO) problem effectively:

  • Machine Learning addresses the scoring of resources and jobs, but alone cannot address the complex combinatorial nature of the assignment problem
  • MIP addresses the complex combinatorial nature of the assignment problem, but alone would require a human to provide a large number of matching scores, making the approach impractical


The Problem

The goal is to find an assignment of resources to jobs that maximizes the total matching score of resources and jobs, while satisfying the requirements of jobs and the availability of resources.

We now dive into the RMO demo itself, starting with a simple scenario. This scenario considers 62 confidential resumes from the Recruitology dataset, four non-confidential resumes, and a matching number of randomly generated jobs. The weights for the job attributes are: {Role: 45%, Technology: 35%, Domain: 20%}. For this scenario, we have disabled some resources and defined some data science jobs as high priority. Also, for another data science job we have set a minimum matching score requirement of 100%. For this scenario, we are allowed to hire a new resource for all jobs, which is the default value.

The “Assignment Quality” KPI (76.2%) measures the average matching score of all the assignments.

Typically, business conditions change frequently and an algorithm that was once efficient for solving the problem no longer works when new business conditions are considered. The new business conditions can be easily considered by adding new variables and constraints to the MIP model and then calling the Gurobi Optimizer to find a solution that recommends the best course of action. This illustrates one of the greatest advantages of using a MIP approach to solve business problems as opposed to using a heuristic.


Access the Resource Matching Optimization Problem

To access the Resource Matching Optimization demo application and create your scenario using your own data from a blank template or to play with existing default scenarios, you must first register for a Gurobi website account and then view the demo.

Guidance for Your Journey

Gurobi: Always Free for Academics

We make it easy for students, faculty, and researchers to work with mathematical optimization.

Trusted Partners, at Your Service

When you face complex optimization challenges, you can trust our Gurobi Alliance partners for expert services.

We’ve Got Your Back

Our global team of helpful, PhD-level experts are here to support you—with responses in hours, not days.

Gurobi Newsletter