ERP systems are a fundamental part of many companies; one property of such systems is a reliance on a Relational Database Management System (RDBMS). Relational databases offer many advantages, which are well researched and documented but also innately possess a rigid architecture. This limitation leads to an ERP system that is inflexible and expensive to modify. Alternatives to relational databases have only seen successful adoption in non-ERP application areas so far.

The work first focused on understanding the goals of the project and the technology vision. This included acquiring the necessary knowledge through reviewing the literature and learning about NoSQL databases. After that, the practical work was structured incrementally. The workload consisted of BPMN modeling, creating an Entity Relationship (ER) diagram and a JavaScript Object Notation (JSON) data model, programming the data generator, defining the queries in the MongoDB Query Engine, measuring and improving query performance, and optimizing the JSON data schemas.

The developed artifacts of this thesis can support the enablerr project with the HR data model, improved event schemas and the evaluation of query performance. They include the generator which can create, upload, export, and measure data with several options to fit specific research scenario needs. The generator can be modified and expanded upon to include more JSON schemas, to broaden its application area. Additionally, query performance analysis results might provide insight into effective data models which can overcome some of the shortcomings of the document-based Database Management System (DBMS) MongoDB.