Motivating Example

To exemplify the complexity of achieving awareness in large-scale control systems and to point out the specic challenges we intend to tackle in this project, we reflect on the area of road traffic management as our demonstration domain. As a motivating example we want to consider the scenario depicted in the following figure, which is charactertic for our real-world setting.

Let us suppose that an operator observes an accident in a motorway tunnel through a surveillance camera, and due to that, a build-up of a traffic jam. The first reaction according to the printed operating procedures would be to open the emergency lane within the tunnel, allowing motorists to bypass the accident and, thereby, avoiding the traffic jam from building up further. Unless consulting a public event information system, the operator might not be aware of the fact that in a short while a football game is about to end and that many spectators will naturally utilize the tunnel for their way home. Although the emergency lane's capacity would not be able to dissolve the traffic jam in this case, spectators could, however, surpass the traffic jam ahead by using a nearby exit. Unfortunately, scheduled roadworks are blocking this exit; a fact the operator is only aware of if consulting the road maintenance time table. Hence, an operator should not only be aware of the option of opening the emergency lane, but also of the non-obvious possibility of canceling the blocking road works before the football fans are stuck in traffic, thus timely avoiding the emergence of a critical situation, i. e., a severe traffic jam. The fundamental problem induced by this scenario is that the overall critical situation has to be derived from multiple heterogeneous information sources (which is of course not as obvious as the magnifying glass in the figure suggests). Furthermore, appropriate actions for this specific case have to be selected, which are most likely not even listed in the operating procedures at hand.