planimbAI was built on the idea that certain individual traits might be influencing how accurately we estimate the time needed to complete our tasks. So we asked people about the timing of their past projects in a comprehensive anonymous survey, and trained a machine learning model on that data in order to capture the relationship between certain traits and how people tend to over- or underestimate the time needed to complete their tasks.
The resulting model is deeply integrated into planimbAI to help you with a more accurate timing of your upcoming projects, leveraging other people’s experiences from their past projects.
Secure Self-Assessment #
For AI timing support to be able to adapt to you, it depends on your inputs to a short self-assessment questionnaire right in planimbAI under Me > AI Timing Support > Start Self-Assessment. It contains a carefully selected part of the questions we asked participants in our training survey. The questionnaire itself is anonymous. It contains only generic questions, the answers to which alone could not be used to identify you, even if someone were to read them in plain text.
Self-assessment works offline without connecting to any servers. Your answers are stored in an encrypted format locally on your device only, and not sent to us or any third-party. They can also be synchronized across your devices via iCloud, so you will just have to complete the self-assessment once. The same goes for reflection inputs in your projects.
Using AI Timing Support #
After you have completed the self-assessment once in planimbAI, you configure a project, estimate its duration and add some tasks. Right from the start, planimbAI’s machine learning model will be there to provide an estimated number of days that a project is deviating from your given schedule, to correct your individual misjudgment (if there is any) on the timing screen. As time progresses, your actual tracked progress will be considered as well for an even more accurate indication of your timing.
Limitations of AI Timing Support #
Based on your self-assessment, AI timing support can only provide an estimated number of days that a project is deviating from your given schedule, and no more. In particular, AI timing support is not able to predict on its own the absolute time needed to complete your projects. The machine learning model used for AI timing support was trained based on a comprehensive statistical survey. As such, it may be accurate for many users, but not for everyone. Therefore, performance and accuracy of AI timing support are not guaranteed.