X-ray vision for procurement

Process mining with process.science expands business intelligence to include the dimension of process flows.In the case described here, the customer started a central pilot project to explore the possibilities ofcombining process mining and business intelligence for the procurement process (P2P).
To his surprise, there was potential for improvement right from the start.

Rapid implementation

process.science relies on the rapid implementation of PoCs or initial projects ‍in order to convey the potential of process mining to all those involved through concrete and rapid improvements to the available tools. In this case, there was an opportunity to expand existing dashboards and existing connections to source systems that the customer was familiar with.
The project was implemented in less than 4 weeks and led to several follow-up projects in which further tasks from controlling and process planning are implemented.

The Goals of the analysis in this case were:

• Identification of savings potential throughpayment at the right time (e.g. discount)
• Identification of current cases that run therisk of missing this goal
• Comparison of expenses for invoice verificationand prevented incorrect payments
• of additional expenses bycorrecting (master) data

Results:

• In the previous year under consideration, a high possible discount amount was not realized
• On average, there is a two-digit percentage of the payment transactions a few working days before the last possible date, on which a discount can still be taken
• The processes were corrected quite often for the participants afterwards
• Impossible event combinations due to existing export errors in the BI
• nonsensical sequences due to configuration errors in source systems

Measures:

• A daily listing of all payment processes that are at risk of not being completed on time led to the result that in the first 3 months since the system was introduced, all unrealized discounts were reduced to almost €0
• Correction of export errors to increase data and forecast quality
• Change of the order and change of the permitted data types of some form fields in order to avoid additional work due to incorrect entries
• Discussion of the process deviation with those responsible to change the existing processes or to eliminate these shadow processes