A Scheduler for Actions with Iterated Durations
Authors: James Paterson, Eric Timmons, Brian Williams
AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we present empirical results to show an improvement in runtime of more than an order magnitude over current schedulers and framing a LTPP as a MINLP. Empirical Validation To validate our approach, we compare BOUNDSEARCH to a TCSPP encoding using a best-first component STP search (COMPSTP), as well as to a MINLP encoding using SCIP as a solver (Achterberg 2009). |
| Researcher Affiliation | Academia | James Paterson, Eric Timmons and Brian C. Williams Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 32 Vassar Street, Building 32-224, Cambridge, MA 02139 {paterson, etimmons, williams}@mit.edu |
| Pseudocode | Yes | Algorithm 1: DOMAINFILTER and Algorithm 2: BOUNDSEARCH |
| Open Source Code | No | The paper does not provide any links or explicit statements about the release of open-source code for the described methodology. |
| Open Datasets | No | The paper mentions generating random data for benchmarking ('The overall temporal constraint was generated randomly', 'A non-linear global preference function was created by generating a random expression tree'), but does not provide concrete access information (link, DOI, formal citation) for a publicly available or open dataset. |
| Dataset Splits | No | The paper discusses benchmarking on randomly generated problem instances but does not provide specific dataset split information (percentages, sample counts, or citations to predefined splits) needed to reproduce the data partitioning. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'SCIP as a solver' but does not provide a specific version number for it or any other software dependency. |
| Experiment Setup | Yes | The LTCs are each of the form: {N [5, 20]; δ [5, 10]; f(N) = a.N}, where a is a random number between 0 and 5. The overall temporal constraint was generated randomly to ensure the solution is varied throughout the state space of possible solutions over multiple runs. Simple temporal constraints (STCs) denoting non-looping activities were randomly interleaved between event pairs with probability 0.2. A non-linear global preference function was created by generating a random expression tree, combining pairs expressions using + and operators with equal probability. |