<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<item>
  <id>05829537</id>
  <dt>j</dt>
  <an>05829537</an>
  <augroup>
    <au>Boudhar, Mourad</au>
    <au>Tchikou, Hamza</au>
  </augroup>
  <ti>Scheduling with arranged multi-purpose machines.</ti>
  <so>Int. J. Oper. Res. 8, No. 3, 379-397 (2010).</so>
  <py>2010</py>
  <pu>Inderscience Enterprises, Gen\`eve</pu>
  <lagroup>
    <la>EN</la>
  </lagroup>
  <ccgroup>
  </ccgroup>
  <utgroup>
    <ut>makespan</ut>
    <ut>complexity</ut>
    <ut>branch and bound</ut>
    <ut>scheduling theory</ut>
    <ut>MPMs</ut>
    <ut>multi-purpose machines</ut>
    <ut>heuristics</ut>
  </utgroup>
  <cigroup>
  </cigroup>
  <ligroup>
    <li>doi:10.1504/IJOR.2010.033545</li>
  </ligroup>
  <abgroup>
    <ab>Summary: In the present study, we consider a special case of multi-purpose machines (MPMs) in which there is a linear order given for the machines. In addition, for each job $J_i(1\leq i\leq n)$, a ``first permissible'' machine $h_i (1\leq h(i)\leq m)$ is given on which it can be processed. Thus, machines $M_{h_i},\ldots , M_m$ are capable of processing job $J_i$ while machines $M_1,\ldots , M_{h_i - 1}$ cannot process job $J_i$. Each job $J_i$ requires a time $p_i$ and the goal is to minimise the makespan. We prove the NP-hardness of the general problem and present some polynomial sub-problems. Heuristics with an exact algorithm of branch and bound type are also presented with numerical experimentations.</ab>
    <rv></rv>
  </abgroup>
</item>