File format

Problem — .ipps

Four sections.

1. Header

<num_jobs> <num_machines> <num_operations>

2. Graph (out)

Starts with a literal out line. Each subsequent line declares the outgoing edges of one operation:

<u> <v1> <v2> ...

means the edges u → v1, u → v2, … . An OR-connector is written by wrapping the OR-set in parentheses with no spaces:

1 (15,11,13)

means operation 1 has an OR fork — exactly one of {15, 11, 13} will eventually be picked.

3. Join (in)

Starts with a literal in line. Each line names the operations that join into one downstream operation:

4 (2,3)

means operations 2 and 3 are the tails of two OR branches that join at operation 4.

4. Processing time (info)

Starts with a literal info line. Three kinds of rows:

• <op_id> start — the job's super-start node (no processing).
• <op_id> end — the job's super-end node (no processing).
• <op_id> supernode — an internal OR helper node (no processing).
• <op_id> <n> <m_1> <t_1> <m_2> <t_2> ... — operation op_id can be processed on n machines; machine m_i takes time t_i.

Minimal example

A 1-job / 2-machine illustration:

1 2 5
out
0 1
1 (2,3)
2 4
3 4
in
4 (2,3)
info
0 start
1 1 1 3
2 1 2 5
3 1 1 4
4 end

The full bundled example lives at data/example/problem.ipps and is rendered below:

Solution — .ippssol

<total_makespan>
<op_id> <ma_id> <job_id> <start_time> <end_time>
<op_id> <ma_id> <job_id> <start_time> <end_time>
...

First line is the total makespan; every subsequent line is one scheduled operation.