De Econometrist neemt een statistische kijk op de wereld.
I was searching for a topic for my thesis, when the idea of investigating the performance measures of a court appeared. This was ideal since one of my relatives is a judge and I therefore could collect many information via him. As it turned out, there were some very important processes that can be optimized to save a lot of resources!
Courts are tied to rules of the government regarding the time a case spends in a court. Yet, there is still space for their own policy. In this policy, they have to deal with some factors, most of the time meaning that the cycle time should be as low as possible. Therefore, it is of great interest whether the work in a court can be done more efficiently.
To investigate whether the work in a court can be done more efficiently, we first want to find out how much time before a hearing the judge and legal assistant should start working on the cases. However, what is most efficient depends on some factors. Factors which can be influenced by a court are the percentage of loss, the time to review a case and the percentage of return on the cycle time. These factors are of interest to decide how long before the hearing a case should enter to process. If a case enters the process six weeks before the hearing takes place, there is a higher chance that the case will be withdrawn in the meantime and the time spent working on the case is wasted than if a case enters the process two weeks before the hearing. However, the three factors mentioned above have different effects on the cycle time. To minimize the percentage of loss and the time to review a case (the time it takes the judge and legal assistant to remember the cases if the cases pass them for the second or third time) it is better to start the process as late as possible, while for the percentage of return (the cases that cannot be discharged after the hearing) it seems better to start the process longer before the hearing, since the chance that the legal assistant and judge have not enough time to gather additional information that is needed is then smaller. Combined with the fact that the involved parties want the case to be finished as soon as possible, this raises the question how the trade-off between these factors should be made.
To answer this question, a model should be made which shows the process of cases going through a court. The figure below shows this model.
When a case is brought to the court it arrives at the administration. Here, all the information that is needed from both parties is collected and the case is assigned to a hearing with five other cases, so a batch of six cases is made. If the batch is complete and assigned to a hearing, the six cases will be discussed in a short meeting by the legal assistant and the judge (arrow ). At this moment in time, the process really starts. After this meeting, the legal assistant will start preparing the cases (arrow a). The legal assistant prepares the cases for the judge, besides checking some things and gathering some more information. If the legal assistant has finished preparing the cases and all the extra information is available, the batch of cases is passed on to the judge (arrow d) and the judge will prepare the cases as well. If the judge still needs additional information for some of the cases, the corresponding cases go back to the legal assistant (indicated by arrow e), who then gathers this information. When all information the judge needs is available and the judge has finished his preparation, the cases will be waiting until two days before the hearing takes place (indicated by the triangle in the figure). Two days before the hearing first the legal assistant and then the judge review the cases (arrows g and d) after which the hearing takes place (arrow h).
After the hearing, the legal assistant and judge have a meeting where they discuss the results of the hearing and how the judgement will be written down (arrow i). Again, the batch is passed on to the legal assistant, who starts writing the judgment for the cases. Subsequently, the judge may add something to the judgment and/or may adjust it, after which the cases are finished and leave the system (indicated by arrow k). It may happen that the parties come to a compromise during the process. If this happens, the case is withdrawn from the process (, and in the figure). It is also possible that some of the cases cannot be dismissed if some extra information is revealed during the hearing, so that further investigation has to take place. If a case cannot be dismissed, it should undergo the process again.
One can argue that the system described above consists of two processes. The first one is the process before the hearing and the second one the process after the hearing. The moment a case enters the system and the moment a case leaves the system can be influenced and should be analysed according to the process before and after the hearing, respectively. How long before the hearing the involved people start working on the case, can have a great influence on the cycle time of a case in a court. If they start working on a case six weeks before the hearing, they can be sure that there is enough time to gather all necessary information in time, so that after the hearing all cases can be dismissed. However, the chance that cases will be withdrawn in the meantime and that the time already spent on a case for preparing it becomes wasted is higher. Furthermore, the time that the legal assistant and the judge need to review the case short before the hearing increases. On the other hand, if they start working on a case two weeks before the hearing, it might be possible that there is not enough time to gather all necessary information. However, in this case the chance that cases will be withdrawn in the meantime is less, as well as the time to review the cases.
The influence of the percentage of loss, the time to review a case and the percentage of return on the cycle time are analysed separately and together. For the percentage of loss, it turned out that the cycle time decreases with an increase in the percentage of loss. Depending on the moment the withdrawal occurs, the influence changes, so four cases were analysed.
What the influence of the different cases is on the cycle time can be seen in the following figure.
Depending on how long before a hearing a case enters the system the service time for a case can be different, since the time to review the case is different. This implies that the cycle time is influenced by the time it takes to review the case for the legal assistant and the judge. And this in turn is influenced by the number of weeks before the hearing a case enters the system. In the figure below, one can see the percentage change in cycle time if the moment of release (in weeks) of cases before the hearing is changed.
One can see that with an increase in the moment of release before the hearing (and thus with an increase in the time to review a case), the cycle time increases as well.
The last variable which influences the cycle time is the number of files that cannot be dismissed after the hearing and should undergo the whole process again. It turns out that the cycle time increases with an increase in the percentage of return (see the figure below).
The above three cases are all cases to better understand how the cycle time changes with changes in the different variables. However, it is much more realistic to combine all variables and see which impact this has on the cycle time. The figure below shows this.
One can see that the cycle time is not monotonically increasing or decreasing, while the cycle times in the figures for the three separate cases are monotonically increasing or decreasing. This means that different variables may have larger or smaller effects if the moment of release changes. Since the graph above is overall decreasing, it follows that the percentage of loss and the percentage of return together have a greater impact on the cycle time than the fact that the legal assistant and the judge need more time to refresh upon a case if a case enters the court longer before a hearing. For example, one may expect that if a case enters the system four, five or six weeks before the hearing, the percentages of loss and return have a greater impact on the cycle time then the time to review a case, since these graphs are monotonically decreasing over time, while the graph of the time to review a case is monotonically increasing over time.
With the input used for this model the shortest cycle time is obtained for a case entering the system six weeks before the hearing. The outcome from this research shows that courts can do quite a lot to change the cycle time of cases in a court. And since a longer cycle time per case means more money spend per case for a court, courts better should have look at possible optimizations.
This article is written by Lotte Post.