The time study we carried out to gather estimates on the process times was limited due to the time constraints we faced. Because of these time constraints, certain aspects of the process were glossed over instead of receiving the attention they required. The VAS toolís performance would improve if the data that was input into the program were more accurate. These include:
∑ The vanís travelling speed was estimated from knowing the trucks average travelling speed.
∑ A per-person catching rate was used in the VAS tool rather than considering the fact that the per person rate changed depending on the total number of people catching. When the litter on the shed floors was in bad condition, the catching rate was slower than normal. This could be incorporated by reducing the catching rate when in bad litter conditions.
∑ The path that the vans travelled along to get between the farms was found by measuring the distance between each farm and increasing the distance by 20%. Although we believed this gave reasonable estimates, the correct distances could be found by having the van record the actual distances.
∑ An average travel rate was found for the trucks. This could be improved by having the travel speed dependent on the roads travelled along or the time of day travelling.
Currently, at the beginning of the shift each catcher is assigned to either the van or one of the trucks. Once assigned, they cannot travel on any other vehicle. This is quite restrictive in comparison to the actual situation. By allowing catchers to travel in more than one vehicle during a shift, the VAS tools validity would be improved. Another aspect not reflected in the VAS tool is that some of the workers do not work for the whole shift.
Lastly, the truck and trailer can be separated at farms, yet this is not considered by the VAS tool. When a truck/trailer split occurs the number of modules, and the down time are affected, and so by considering truck/trailer splits the scheduling would be improved.