Discover opportunities for improvement with process mining

Discover opportunities for improvement with process mining

Insurers have an enormous amount of data. They only use part of this data however to optimize processes, especially data they know is predictive of the damage to be paid. In the case of an occupational disability or life insurance policy, this is construed from the medical history of the customer. In the case of car insurers, they use number of damage-free years, brand, type of car and age of the policyholder. And in the case of a building and fire insurance policy, it is the rebuilding value of the house and whether or not the house has a thatched roof.

Better forecasting is better anticipation
In addition to predicting the chance of damage, there are dozens of other processes you could possibly improve if you had more knowledge about how this process actually works. Are there, for example, damages that occur periodically due to weather conditions? Can you predict in advance that there’ll be a spike in requests for offers for certain insurance policies? For this You-Get uses IBM Watson. The better you can predict each subsection of a process, the more efficiently you can set it up to anticipate changes.

Discover possibilities for improvement
With analytics it is possible to analyze quantitative information about processes. An example are the ‘event logs’ that make almost every software package. It indicates when which action took place. While these log files were not available in the past, this information is now fairly easy to add to a ‘data lake’. By analyzing large amounts of quantitative data you gain insight into how processes actually run to better identify bottlenecks. Examples include deviating process paths, delays, lead time overruns, hidden inefficiencies and rework. In other words, process mining provides insight into what’s really going on.

Monitor result improvement action
By using process mining it is also much easier to monitor the result of a process improvement action. The data show immediately whether the lead time is shortened and the degree to which rework decreases. It is also easy to objectively compare similar processes in two locations. You can not only benchmark the results of different branches, but also the processes. Thus, you immediately see where the differences between branches come from.

Do you want to look in a fact-based way for improvement opportunities in your processes, and don’t know where to start? Contact us with your challenge.

Caspar Pille, Cognitive specialist at You-Get

+31 (0)20 737 02 76