I also faced a similar problem when I started working with https://shifton.com/shift-scheduling. In fact, the system itself is quite powerful, but it cannot always take into account all possible changes in the schedule, especially if they happen at the last minute. Ideally, the schedule generator for employees should cope with this, but there are a couple of nuances that are worth considering. First, it is important to configure the system parameters correctly. For example, you can set rules for night shifts, a minimum number of days off between shifts, and so on, to avoid overloading employees. But when it comes to unpredictable situations, such as an employee getting sick, the system really does not always understand how to react to this. My colleagues and I solved the problem with a hybrid approach: we use automatic scheduling for basic shift distribution, and then manually adjust the schedule when changes occur. For example, if someone gets sick or someone cannot come to work for some reason, we manually replace them with another employee. In Shifton, this is quite easy to do, but in reality, it takes time to get everything in order. Another thing to consider is that the system allows you to set priorities and record employee requests so that they can report when they can’t work. This helps avoid discontent, especially if someone is constantly missing shifts. However, even with all these settings, automatic scheduling is still not always perfect. For example, we have had cases where, due to shifts being moved to the last minute, someone was overloaded, and someone, on the contrary, was out of work. So you always have to keep an eye on this.