Two traffic assignment codes were developed - named the CAON Code and the IAON Code.  Both were designed to complete an MSA traffic assignment on a network and had an all-or-nothing assignment core.  Both codes were tested on three small, hypothetical networks.  A four node, 13 node and 25 node network were the networks used for testing.  The results of testing both the MSA and AON of each code on the 13 and 25 node network are seen next to one another in the series of images below. 

​

The color of the lines represent the volume/capacity ratio on each link.  Red represents a congested link with a high volume/capacity ratio, between 1 and 0.66.  Yellow represents an averagely congested link with a volume/capacity ratio between 0.66 and 0.33 and green represents an uncongested link with a volume/capacity ratio between 0 and 0.33.  The thickness of each line also represents the number of vehicles on the link.  

​

As is evident from the images below the AON assignment restricts the flows to the shortest paths while the MSA assignment distributes traffic across the network evenly.

 

While there were discrepancies in the volumes of traffic assigned to each network between the CAON and IAON codes both codes had the same bheaviour in creating an evenly distributed traffic flow.  The most telling result is that there are much fewer red links in each of the networks after running the MSA code.  

​

Only the CAON code was tested on the City of Cape Town Bus network as the IAON code ran into extensive memroy errors due to the size of the network.  It took  the CAON MSA code two hours and 40 minutes to find a convergence.