Problem instances for outbound truck loading and scheduling problem

<p>The following dataset contains problem instances for the outbound truck scheduling and loading problem, which are proposed in the work &quot;Giorgi Tadumadze &amp; Simon Emde (2021): Loading and scheduling outbound trucks at a dispatch warehouse. <em>IISE Transactions</em>, DOI: <a href="https://doi.org/10.1080/24725854.2021.1983923">10.1080/24725854.2021.1983923</a>&rdquo;.</p> <p>The problem instances are stored in table &ldquo;instances&rdquo;, where columns of tables can be interpreted as follows:</p> <ul> <li>ID: &lt;running index&gt;;</li> <li>name: &lt;instance name, specifying the number of items <span class="math-tex">\(m\)</span>, number of trucks <span class="math-tex">\(n\)</span>, value of parameter <span class="math-tex">\(\alpha\)</span>, [value of parameter <span class="math-tex">\(\Delta\)</span>], and the width of trucks&rsquo; time windows&gt;;</li> <li>O: &lt;number of served OEMs&gt;;</li> <li>m: &lt;number of items&gt;;</li> <li>n: &lt;number of trucks&gt;;</li> <li>Q: &lt;total number of available workers&gt;;</li> <li>D: &lt;total number of available dock doors&gt;;</li> <li>w_i: &lt;vector with <span class="math-tex">\(m\)</span> elements: the <span class="math-tex">\(i\)</span>-th element corresponds to the size (required space) of item <span class="math-tex">\(i\)</span>&gt;;</li> <li>d_i: &lt;vector with <span class="math-tex">\(m\)</span>&nbsp;elements: the <span class="math-tex">\(i\)</span>-th element corresponds to the deadline of item <span class="math-tex">\(i\)</span>&gt;;</li> <li>r_i: &lt;vector with <span class="math-tex">\(m\)</span>&nbsp;elements: &nbsp;the <span class="math-tex">\(i\)</span>-th element corresponds to the relative importance (penalty cost per time unit of earliness) of item <span class="math-tex">\(i\)</span>&gt;;&nbsp;</li> <li>c_j: &lt;vector with <span class="math-tex">\(n\)</span>&nbsp;elements: the <span class="math-tex">\(j\)</span>-th element corresponds to the capacity of truck <span class="math-tex">\(j\)</span>&gt;;</li> <li>a_j: &lt;vector with <span class="math-tex">\(n\)</span>&nbsp;elements: &nbsp;the <span class="math-tex">\(j\)</span>-th element corresponds to the earliest possible departure time of truck <span class="math-tex">\(j\)</span>&gt;;</li> <li>b_j: &lt;vector with <span class="math-tex">\(n\)</span>&nbsp;elements: the <span class="math-tex">\(j\)</span>-th element corresponds to the latest possible departure time of truck <span class="math-tex">\(j\)</span>&gt;;</li> <li>q_i: &lt;vector with <span class="math-tex">\(m\)</span>&nbsp;elements: the <span class="math-tex">\(i\)</span>-th element corresponds to the number of required workers to prepare and load item <span class="math-tex">\(i\)</span>&gt;;</li> <li>rho_i: &lt;vector with <span class="math-tex">\(m\)</span>&nbsp;elements: the <span class="math-tex">\(i\)</span>-th element corresponds to the handling time of item <span class="math-tex">\(i\)</span>&gt;;</li> <li>B_i: &lt; <span class="math-tex">\(m \times n\)</span> matrix: each entry in <span class="math-tex">\(j\)</span>-th column and <span class="math-tex">\(i\)</span>-th row corresponds to the binary parameter which has a value 1 if set of available trucks <span class="math-tex">\(B_i\)</span> contains truck <span class="math-tex">\(j\)</span> (i.e., if truck <span class="math-tex">\(j\)</span> departs towards the OEM, who ordered item); 0 otherwise&gt;;</li> </ul> <p>The first 270 entries (ID between 1-270) contain OTSLP instances with different instance sizes, used for the computational performance experiments (Section 5.1).</p> <p>The following 100 entries (ID between 271-370) contain 40 OTSLP instances with the varying time window width for each truck (ID between 271-310), 30 OTSLP instances with the varying level of available workers &nbsp;(ID between 311-340), and 30 OTSLP instances with the varying level of available dock doors &nbsp;(ID between 341-370), used for the managerial inside experiments (Section 5.2).</p> <p>The detailed computational results for each instance and solution approach are reported in tables, which are named with the following convention: &lt;results_&lt;approach&gt;&rdquo;. Specifically, we report the required computational runtime in CPU seconds, status of the found solution (&ldquo;Optimal&rdquo;, &ldquo;Infeasible&rdquo;, &ldquo;Feasible&rdquo; / &ldquo;AbortTimeLim&rdquo;), as well as the best found upper (and lower) bound in columns &ldquo;runtime&rdquo;, &ldquo;status&rdquo;, &ldquo;UB&rdquo; and &ldquo;LB&rdquo;.</p>