Robot process automation in the WOZ objection (chain) process

Municipality of Tilburg

To support the performance of administrative operations in the WOZ (chain) process, algorithms are used. Robotic Process Automation (RPA) is the method we apply. In RPA, repetitive, rule-based operations are automated using algorithms.

Last change on 26th of June 2024, at 19:03 (CET) | Publication Standard 1.0

• Publication category: Other algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

• Economy
• Public finance

Begin date

2023-05

Contact information

[e-mail]

Responsible use

Goal and impact

The aim is to reduce the processing times of WOZ objections using RPA, so that citizens receive a decision on their objection faster.

We also want to automate repetitive and time-consuming employee processes to (continue to) manage the increasing workload.

The algorithm supports employees in their tasks. The executive employees of the Municipality of Tilburg get help with administrative and repetitive tasks, giving them more time for substantive work, experiencing less workload and stress, and more job satisfaction.

Moreover, the algorithm helps reduce the need to hire external staff.

Considerations

Trade-offs: To provide timely feedback to the citizen/business related to legal deadlines, it is necessary to provide automation support to the work process.

RPA is ideal for this as the current level of privacy, security and information management is not affected. All personal data is processed encrypted, ensuring data security. An alternative to relieve staff and speed up the process would be to hire skilled external staff. However, this is very costly and given the tightness in the current labour market, it is not always an available option. In addition, RPA is scalable and flexible, making the WOZ objection process more robust against large annual fluctuations in the number of objections.

This robot thus reduces social costs.

Human intervention

The algorithm is currently deployed to support/relieve employees in the administrative work within the WOZ (chain) process. This means that the algorithm automatically fills in information from objections in the municipal systems. In addition, the algorithm links a code (grievance) to each objection reason. This allows officials to directly access the information they need to handle objections. The appraiser then does the substantive research and thinking related to appraising a property and drafting a decision on an objection himself.

When the algorithm cannot process a file or cannot process it completely, it is offered to the employee to examine it and manually process the administrative steps. Depending on where this is in the process, this is a customer contact employee or a legal assistant. This person picks up the file from then on. It can also be chosen by the employee to offer the file again to the robot for the follow-up process after this.

Risk management

Risk management: Risks in deploying the algorithm have been tackled because it uses quality checks. If a citizen/object is not recognised or cannot fully process an objection letter to relevant objection reasons, the cases fail. These are then presented to an employee to review and process manually.

Legal basis

Law and regulations on property valuation

Operations

Data

NAW data

WOZ data

Technical design

From two different channels (web forms and mailbox), WOZ objections from citizens and NCNP offices arrive at the algorithm. The robot enters the objections into the appropriate applications. Next is the next robot that de-grids (links the objections to objection code) and links each objection reason to the correct grievance (code). This is followed by two more administrative robots that take over administration from legal assistant and the appraiser in different applications.

Technical operation: Robotic Process Automation (RPA) is the method we use. In RPA, repetitive, rule-based operations are automated using algorithms. These algorithms can perform tasks such as data entry, processing and movement, eliminating human intervention and significantly increasing efficiency. Only rule-based code is used. Thus, there is no intelligence embedded in the algorithm.

External provider

Internally developed