Cement Americas

WIN 2016

Cement Americas provides comprehensive coverage of the North and South American cement markets from raw material extraction to delivery and tranportation to end user.

Issue link: https://cement.epubxp.com/i/636168

Contents of this Issue

Navigation

Page 13 of 39

12 CEMENT AMERICAS • Winter 2016 • www.cementamericas.com FEATURE what you say. The LP is quite literal; it does exactly as we say and not as we mean. So therefore, analyzing the answer to understand the nature of the underpin- ning moves is essential in this fnal phase. As diligent as we intend to be on the front end, it is very easy to overlook an obvious constraint, which can wreak havoc on the answer. For example, a plant might fnd it very cost effective to make product and then keep it all to itself in order to minimize shipping costs – unless con- straints are in place to make this a very unattractive (or impossible) option. This phase may take a few runs to ensure the model is error-proofed. So what picture emerged from the abyss? The most proftable, optimized scenario shows a signifcant shift in manufacturing sourcing to Plant C at the expense of Plants A, B and D (Exhibit 3). These sourcing shifts are motivated from the identifcation of more effcient distribution modes in Markets 1-6 (Exhibit 4). The larg- est swing is in Market 3, with Plant C trading nearly 70 percent of its market volume to Plant D. These moves translate to a savings of $2.1 million in this one region (while still meeting customer requirements.) And from a logistics standpoint, one can see the forecasted shift in transport needs during September-October from rail to truck (Exhibits 5-6). This shift effectively allows the company to roll off excess railcar leases and switch to the more variable delivery cost in trucks. Exhibit 3: Shifts in product supply shifts. NETWORK DEBOTTLENECKING Every process has a limiting factor such as market demand, funds, equipment capacity, time, etc. that prevents perpetual fow. And not all bottlenecks are created equal. Some may have no value since they are captive to something more critical while others are un- doubtedly costing the enterprise is sales or excess labor. Exhibit 4: The impact of product supply on the deliv- ery routes. How does one know which ones are worth resolving and the respective paybacks? Networks are not unlike the timeless Rubik's Cube toy – dynamic and intertwined – and fguring out the overall impacts of knocking down a bottleneck is something easily accomplished with a program like this one. The team was able to view processes in the network that were "maxed out" (or at 100 percent) by setting up output tables derived directly from the results of the LP. As the example shows in Exhibit 7, Plant B was potentially holding up the works as it was forecasted to be loading product on trucks and railcars May through October 24 hours a day. Though there are more technical ways to measure the bottleneck's value and signifcance, the team found it was a relatively simple exercise to incrementally raise the throughput level at Plant B in the model and study the changes overall proft return. In this case, there was no signifcant return since there were other limiting con- straints in the downstream storage capacities. When the storage capacity issue was isolated and "fxed" in the model, the result was like unclogging a drain: volume fowed much easier. Without network modeling ability, it would have been very diffcult to quantify these critical bottlenecks and their accompanying values.

Articles in this issue

Links on this page

Archives of this issue

view archives of Cement Americas - WIN 2016