This repository contains the problem and solution data used for the experiments in the paper. Moreover, the algorithm MAPLE script is included.
The problem instance files follow the following format.
- The single-learning case (PSP-SL)
(Note that the instance files contain resource information which is not used in our models.)
% (NbTasks) = (number of project jobs); NbTasks = 32;
% (Tasks) = {<(job),(original duration),(Unused resource demands),{(successor1, sucessor2, ...),[(teacher job,reduced duration)]}>, ...}; % Note that teacher and alternative duration are set to zero if not a learning candidate. Tasks = { // < Task ID, Duration, [Resource Demands], {Successor(s) (optional)}, [Teacher, Alternate Duration] > < 1, 0, [0, 0, 0, 0], {2, 3, 4}, [0, 0] >, < 2, 3, [9, 0, 9, 0], {17, 25, 29}, [22, 2] >, < 3, 5, [0, 10, 3, 0], {9, 10, 16}, [0, 0] >, < 4, 9, [6, 2, 2, 0], {5, 6, 7}, [2, 8] >, < 5, 10, [0, 0, 10, 1], {17}, [0, 0] >, < 6, 1, [10, 1, 2, 0], {13, 16, 22}, [0, 0] >, ...
- The multi-learning case (PSP-ML)
% (n) (number of jobs) n 122 % (job) (duration) (reduced duration when learning) j d dAlt 1 0 0 2 8 4 3 5 5 4 8 8 5 1 1 6 8 4 7 2 1 ...
% (mA) (number of precendence arcs) mA 183 % (arc tail job) (arc head job) 1 2 1 3 1 4 2 5 ...
% (mL) (number of base learning opportunities) mL 30 % (arc tail job) (arc head job) 2 44 5 31 10 101 17 36 ...
% Multi-learning data LL % (L) (base learning opportunity; from above) (number of learning scenarios) L 2 1 % (t) (number of teachers in this scenario) (teacher1) [(teacher2) ...] t 1 79 % (d) (learner duration in this scenario) d 4 L 6 1 t 1 31 d 4 ...
The solution (project schedule) format (both instance types) is as follows. % (job),(start time),(end time) 1,0,0 2,0,6 3,0,5 4,0,9 5,9,10 6,10,13 7,6,13 8,10,13 9,6,9 10,13,20 ...