README

• Generates random numScores, partners, adjMatrix, itemWidths, allItems (numbering each item)

• Creates population of solutions, one using FFD, the rest using FFR

1. Straight swap, no moveType
   • stripX = 2 items and stripY = 2 items
2. moveType = 11
   • stripX = 2 items and stripY > 2 items, c and d adjacent
3. moveType = 0
   • stripX = 2 items and stripY > 2 items, c and d not adjacent
4. moveType = 12
   • stripY = 2 items and stripX > 2 items, a and b adjacent
5. moveType = 0
   • stripY = 2 items and stripX > 2 items, a and b not adjacent
6. moveType = 0;
   • stripX > 2 items and stripY > 2 items

1. Straight swap, no moveType
   • stripX = 2 items and stripY = 1 item
2. moveType = 21
   • stripX = 2 items and stripY > 1 item
3. moveType = 22
   • stripY = 1 item and stripX > 2 items, a and b adjacent
4. moveType = 0
   • stripY = 1 item and stripX > 2 items, a and b not adjacent
5. moveType = 0;
   • stripX > 2 items and stripY > 1 item

1. Straight swap, no moveType
   • stripX = 1 item and stripY = 1 item
2. moveType = 31
   • stripX = 1 item and stripY > 1 item
3. moveType = 32
   • stripY = 1 item and stripX > 1 item
4. moveType = 0
   • stripX > 1 item and stripY > 1 item

1. moveType = 41
   • stripY = 1 item
2. moveType = 42
   • stripY = 2 items
3. moveType = 0
   • stripY > 2 items

• Choose two solutions from population, these will be parents (parent1 = stripX and parent2 = stripY)
• Calculate fitness of each of these parent solutions (parent1Cost and parent2Cost using Fitness function)
• Run GGA/GPX
• Go through strips in offspring, find strips that have best fullness, add these to qualityStrips
• Calculate fitness of offspring (offspringCost using Fitness function)
• If parent1 has worst fitness, remove parent1 from population and replace with offspring
• Else if parent2 has worst fitness, remove parent2 from population and replace with offspring
• Else if parent1 and parent2 have equal fitness, choose at random.
• If fitness of offspring is better than bestFitness (current best fitness of a solution in the population), set bestFitness = offspringCost
• End EA

• Choose two strips k and l from stripY randomly
• Mark all items in strips between k and l (including k and l) as checked
• Go through stripX, if a strip contains an item that is checked, delete entire strip from stripX
• Put strips between k and l (including k and l) from stripY into stripX
• Go through stripX, mark all items as checked
• If all items are in stripX, offspring = stripX, return to EA function.
• If any items are not in stripX (i.e. checked == 0), put these items in absentItems
• Clear stripY, run PartialFFD on absentItems and stripY, then run LocalSearch using stripX and stripY, return to EA function.

• Choose fullest strip out of all strips in stripX and stripY (e.g. fullest strip is in stripX)
• Mark items in fullest strip as checked, put this strip in offspring, erase this strip from stripX`
• Go through all strips in stripY, if there is a strip that contains a marked item, delete entire strip from stripY
• Now go to stripY and choose fullest strip
• Repeat, alternate between stripX and stripY
• If all items are in offspring, do nothing (i.e. absentItems vector is empty), return to EA function
• If any items are not in offspring (i.e. checked == 0), put these items in absentItems
• stripX = offspring, clear offspring, clear stripY, run PartialFFD on absentItems and stripY, then run LocalSearch using stripX and stripY, return to EA function.