Automatic invoice matching – Van den Bosch



Company Name / Department Van den Bosch
Contact Person Teun Franssen
 Location Erp
Optional remote work Yes, at least two days / week at the office
Travel expenses (own account or reimbursed by the company) Reimbursed by the company
Housing arranged by company No
Housing expenses (how much per month, own account or subsidized by the company) Own account
Internship compensation  €500,- per month
Study program
ESCF community


Start date

February 2024


Company Description

Van den Bosch is an international logistics service provider located in Erp, The Netherlands. They aim to drive sustainable and reliable change in the supply chain through intelligent innovation. More than 80% of all their transport is performed by means of intermodal transport. They believe that logistics is not just about transport, but also about data, insights, and know-how. They are constantly trying to drive change within the sector, which is why they are known as “The Supply Changer in Bulk”.


Project Description

Project description:

We receive approximately 2000 invoices per week from our suppliers. 70% of these invoices is based on an electronic order (Purchase Order, PO) which is sent to the supplier. If the invoice amount matches the PO amount, the invoice is matched automatically.
Unfortunately, around 15 to 35% of these invoices cannot be matched automatically due to various reasons. The differences need to be checked manually and decided what to do per cost line. This is time consuming work, it jeopardizes timely payment to our suppliers an does not fully utilize our system capabilities.
Differences can be categorized in a.o. incorrect prices on the invoices, unallowed extra charges, incorrect masterdata (resulting in incorrect prices on our PO’s) etc..

Goals of the project:

In this project we would like to recognize patterns to increase the matching rate and decrease the amount of manual adjustments per invoice. A thorough investigation is requested to determine the various reasons for mismatches and (automated) ways to resolve these.


    Suggestions on improvements in the invoicing process.


    Essential student knowledge:

    Machine Learning / AI




    More information:  


    logo vanderWal 240x140