Meeting 7

Meeting 7: 21/08/19

Cost Function discussion

• A* cost function quite difficult to implement - assigned to @Vanessa, @Allen, @Richard
• Cost function helps reduce the search space while also producing a optimal output
• TRY TO UNDERESTIMATE COST FUNCTION SO THAT WE DONT LOSE OPTIMALITY
• BottomLevel, DRT, IdleTIme

Pruning Discussion

• Pruning reduces the search space and thus speeds up execution of the A* Algorithm, assigned to @Hong, @Michelle, @Dinith
• 3 Pruning techniques we are implementing?? - Duplicate Detection, Normalisation, Identical Tasks

I/O Discussion

• Possible thought of removing Edge/Node classes and switching to primitive data structures to reduce memory usage @Dinith is looking into it
• Will have to revamp a lot of the code, still deciding

What we have done/close to finishing

• Basic A* implementation which gives optimal output, but is incredibly memory inefficient, unable to process Nodes_11_OutTree
• Every has read the report to get full understanding and scope of the problem