Route Planner Knowledge Source

The purpose of the Route Planner knowledge source is to fix missions which have been rendered inconsistent as a result of introducing changes (as a response to disruptive events). Inconsistencies can be temporal (successor leg departs before predecessor leg has arrived) or geographical (successor leg departs from a different airfield where the predecessor leg ended).

The set of missions initially available is considered to be known beforehand. The route planner will not be used in the generative mode: it does not generate missions to satisfy the patients requirement; it only fixes existing missions if the analyzer identifies that the disruptions introduced inconsistencies affecting the current set of missions. When scheduling patients, the system assumes that all missions in the system are temporally and geographically consistent.

The general behavior of the route planner is to remove conflicting legs from a mission and add new ones. According to the type of conflict, two different strategies are implemented:

• If there is a gap in the sequence of legs, that is, if there is a pair of airfields in the itinerary of the mission that is not connected by any leg, the system will generate a new leg filling the gap and will add this leg to the mission. If the new leg introduces temporal constraint violations, that is, the new added leg finishes before the next leg in the sequence, the remaining legs will be shifted by the required amount. The patients involved in the conflict and identified during the analysis process are allocated to the new leg. This strategy is used, for example, when an intermediate airfield of a mission is closed. In this particular case, the closed airfield will be bypassed and all patients allocated in the arriving leg that got cancelled will be sent to the new destination.
• If there are no gaps in the sequence of airfields served by a mission but the last destination is not available, the last leg of the mission will be removed and a new leg having as destination an airfield as close as possible to the old destination will be introduced. All patients of the cancelled leg are transferred to the new created leg. This strategy is useful, for example, when the final airfield destination of a mission is closed when the aircraft is already en route to the closed airfield.

A third strategy that could be easily implemented is adding new legs to existing missions. This strategy would be useful in the case of resource breakdown requiring emergency landing or to pick up patients in locations not served by any missions. In this case, the user would have to specify insertion points in the mission itinerary for the new legs and respective destinations.

This knowledge source does not deals with patient's conflict. It will perform the default behavior of allocating the affected patients in the added legs. The occurrence of new conflicts will be verified and signaled. The analyzer will be responsible for identifying whether previous conflicts are still valid after the mission has been fixed.