Speed

• Operational speed: the time that elapsed between observed departure times at the stops involved. At the last stop of the route the observed arrival time is used.
• Passing speed: the time used in operational speed, minus the time that elapsed while waiting at stops.
• Company speed: the time used in operational speed, minus any delays that occurred between stops.
• running speed: the time used in passing speed, minus any delays that occurred between stops.

Clearly, operational speed is the lowest and running speed is the highest of these four types.

In the graph, the horizontal axis shows two-letter abbreviations for the halts along the route. The vertical axis shows speed in km/h.
Between each pair of halts there are two red arrows and a gray bar separated in two by a small white line. The red arrows indicate the lowest and highest observed speed. The bar indicates the 15% and the 85% values of the speed; thus 70% of the observed speeds was between the top and the bottom value of the bar. The white line in the bar indicates the mean observed value.
On the right of the graph, the average values over the entire route are shown.

The highest observed speeds are quite miraculous. The explanation for these values is that the on-board computers do not always register the time that a vehicle came to a stop or started moving again. When the engine of the bus was switched off, sometimes truely erratic values are recorded. In the 1191 trips used for this graph, there were obviously a few errors in the recorded times. See the Trip matching page for a list of problems and potential causes in AVL and board computer data.

The mean values and the 70% interval are much more reliable. From the graph it is clear that the operational speed between the first two stops is very low (on average about 17 km/h). The average speed over the entire route is quite good (about 31 km/h). The low average speed between the first two stops is caused by delays between those stops.

It is to be expected that the width of the speed probability distribution is much smaller over the entire route than over individual route sections. In fact, the width of the speed probability distribution is inversely proportional to the length of the route (section).

TRITAPT can also produce a speed graph that shows all observed speeds. In this graph the individual speeds are indicated with fine horizontal green lines. The 1191 trips used for this graph causes many of the green lines to touch, resulting in bigger green areas.

From this graph, one can get an impression of the probability distribution of the speeds that occurr on this route. The data in this graph can also be shown in a tabular format. In this table one can easily locate trips with exceptionally high or low speed.