A process exists only once a deduction answered a situation. The simulations are therefore saved by situation (the actualities and risks active then), task and actor: the next deduction of the same situation finds them again, whatever identifiers it gives. Each group says whether it is Played now or Not played now, and Delete all forgets its simulations. They are not part of the story: the timeline neither shows nor replays them.

• On: checked, the task is simulated and r-iosemit gives its field values when it is accepted (the path of its actor for a move); unchecked, the task is real: its actor types the values, or a real GPS reports the move.
• Auto and Outcome: see below.
• Task, Actor (r-io suite for a wait, run by the workflow engine) and Task status as in the to-do lists, including tasks done before by the preventive procedure and tasks ignored when the corrective took over.
• Simulated and Expected: the indicators of the KPIL script of the project and the outputs of the task. A simulated value is the middle of the expected interval by default and can be changed here, as can an output; a changed value is kept for the next supervision.

A task marked Auto does not wait for its actor: it is accepted as soon as the process reaches it and ends after its field duration, Done, Error or Reject (at once). The clock of the use case is brought to its end first, the steps of the story met on the way played before. A wait of the process ends at its planned end the same way. In the Play box of TIMELINE, Tasks › Auto takes every task at once, with field values As expected, Random, Always below or Always above within a margin, and can keep the values changed in TASKS.

For a move task, Edit chooses what the actor does: the Same path faster or slower, Another path to the same destination, or Gets lost and ends elsewhere, at a pace of 50 to 200 % of the expected one and a speed of ×1 to ×16 for the simulation. The move is played as GPS positions of the actor; an actor without a GPS only has the duration of its move measured. Show on a map draws the expected path and the simulated one.

Once a simulated task is over, the tooltip of its values sent state lists the sensor values it sent at its end: a task that leaves a resource in a state (a barbecue put out) sends the value of the sensor that gives this state, then holds it, as on the field. These task effects are drawn on the row Expected effects from the tasks under the timeline and can make the playback switch tracks (see Forks and tracks).

The field tasks walkthrough of IMT Albi Crisis shows the whole table: an output set by hand, a move played on another path, a task switched off to be done for real. In Orléans Flood Crisis, the closure, inspection and reopening tasks run in Auto and the waits end at their planned end; in the Barbecue, Paul and Julie walk to the barbecue or to the pool and do their tasks on their own.

Reference: Scenario timelines › simulated tasks.

r-iosemit plays the field, not r-io suite. No r-io suite application calls it or knows that it exists: for r-io suite, every event it plays is a real one. It reads the knowledge base of r-io suite (states, positions, alerts, risks, tasks) to choose its tracks and to record its runs, without ever writing what it reads. It acts only through the channels of the field: the sensors (their datasets, planned values, GPS moves and forecasts published as a provider would), the messaging sensors, the human-task service for the simulated tasks, and the clock API of the use case once the operator opened it in r-ioga (Edit project › Use case clock › Allow remote clock control).

A few authoring items act on the knowledge space directly, and the dialogs say so: a planned risk level (like the Potentialities Management slider), a referenced actuality (a FREEZE actuality made ACTIVE, or a clone of a REFERENCE one), Previous step and Restart (they restore a checkpoint of the knowledge space kept by the server) and Replay (it writes the supervision settings of a run back, after a warning).