These datatasets relate to the computational study presented in the paper "Machine-based Production Scheduling for Rotomoulded Plastics Manufacturing", authored by Mark Baxendale, James McGree, Aaron Bellette and Paul Corry. They consist of all the randomly generated problem instances.

Problem Instance Data Format:

The included zip file contains a folder for each problem instance (numbered 0 - 49) associated with each F parameter setting (F0, F0_5, F1) and each number of jobs (30, 40, 50). Each problem instance is described by 5 csv files: jobs.csv, machines.csv, scalars.csv, compat.csv, and machcompat.csv.

jobs.csv - Each row represents a different job. The first four values in each row represent the processing times for the load, cook, cool and unload stages respectively. The fifth column gives the surface area of the job. The sixth column defines the due date.

machines.csv - Each row represents a different machine. The first value represents the number of arms on the machine. The second value represents the surface area on each arm.

scalars.csv - The first value represents the number of jobs. The second value represents the number of machines. The third value represents the number of production stages (4). The fourth value represents the M parameter. The fifth value is the operational time per day (mu). The sixth parameter is the non-operational time per day (nu).

compat.csv - This file provides all sigma values, used to indicate the compatibility of batching two jobs together. The value in the ith row and jth column gives the compatibility of batching jobs i and j together.

machcompat.csv - This file provides all kappa values, used to indicate the compatibility of assigning a job to a machine. The rows correspond to machines, the columns to jobs, where the value in the ith row and jth column gives the compatibility of assigning job j to machine i.