What Causes Rack Loading Disruptions?

November 15, 2013 By

By Dennis Gannaway

The distribution of gasoline and diesel from refineries and import points to sales terminal loading racks is a complex activity. Supply programs are often buffeted by shifting economics and a variety of intrusive factors. Most product supply groups have developed operating practices that prevent supply chain problems from affecting tankwagon customers, but past analysis has indicated room for additional performance improvement.

Bulk transportation of fuels is via truck, train, barge and deep-water vessels, and most significantly through product pipelines. There industry practice is to treat the physical in-transit and in-tank inventories as fungible. Consequently a material change in the shipments of one participant can have an immediate and unexpected impact on other shippers and position holders in the comingled inventory.

Supply plan upsets may originate from operational problems at refineries or pipelines or at bulk storage locations. Weather or other natural conditions sometimes are the disruptive factors. In many markets inventory churn occurs each spring and fall as suppliers scramble to comply with seasonal specification changes. And there are the market-induced changes. It is not uncommon for competition in rack localities to jerk sales volumes up and down, creating a price dynamic unique to that metro market. The cost at origin, typically based on highly variable spot prices for crudes, feedstocks or motor fuels, makes the margins for refinery-to-rack movements even more volatile. Margin fluctuations often motivate opportunistic trading that can sway volumetric inputs and offtakes for the whole supply route.

Experienced supply managers have seen all of this before. Most downstream supply organizations deal with the complexity and the discrete risks of fuels distribution by working within established scheduling patterns, maintaining time-tested inventory targets and relying on numerous contingency arrangements with alternate supply points, alternate suppliers, standby exchanges, etc., when deviations occur. Reciprocity with other shippers and suppliers establishes a cooperative environment conducive to working situational fixes. It pays to be fully informed of current conditions and to maintain good relationships with all (or almost all) of the supply chain players.

Overall it is an effective combination of practices, and most shorts and imbalances are resolved before the distributor’s trucks arrive to take delivery. How effective is it? Several years ago a major downstream company, with fuels sales throughout the U.S., looked at the reliability of their supply network. Their study relied to a large extent on the company schedulers’ incident logs that recorded how often truck loading at sales terminals was delayed or prevented altogether. Initially the log data indicated loading was problem-free 98% of the time. After a re-check indicated the logs were not capturing all loading upsets the problem-free percentage was judged to be about 95%.

That is still good performance. But at 95% a proactive organization will look for ways to make further improvement. While the supply department’s managers can provide a wealth of insight on problem causes and potential cures, they are often engaged in the daily efforts to keep supply moving as smoothly as possible. If appropriate historical data is available, a comprehensive analysis is a better starting point to spot patterns or pervasive trouble factors and to assess their relative weight in affecting performance. In the reliability review mentioned above the incident logs were a rich source of feedback. (The fact that they did not include all interruptions was not considered a systematic error.) They showed the incidence and locations of loading interruptions, the duration of the problems and products involved, and also provided a cursory description of the presumed cause for each such episode.

The breakdown of causal factors, tallied for a year at almost 400 terminals across the U.S., is shown below. The first chart shows the number of loading rack disruptions by category taken as a percentage of all loading rack disruptions. The second quantifies the estimated product impact of each category of loading problems. To do this each loading disruption was weighted based on its duration, the number of products affected and the normal level of sales for each product. As before the product impact for each causal category is expressed as a percentage of total product impacts.



Planned and unplanned maintenance at terminals, as well as snafus occurring in other terminal processes (often involving sales authorization systems) accounted for over half of Disruption incidents. In terms of Product Impacts, terminal-related factors are still significant, adding up to 19%, but that is a lot less than 50+%. Evidently many of the loading problems that arise from activities at the sales terminals do not last long or do not affect that many products.

Pipeline issues were also shown as a major cause of Disruptions (30%) and an even greater contributor to Product Impacts (62%). That is logical considering pipeline delays frequently last for days and affect many if not all products moving into the terminals.

Surprisingly, demand spikes and allocation cut-offs had relatively small effects on product availability. While that was true based on data aggregated to the national level, those factors had much bigger effects at specific terminals. Similarly, the percentage of upsets attributable to Refinery Supply Issues was unexpectedly very low. That is a fair characterization overall, but the downstream effects of refinery events could have been somewhat underreported. Many refinery problems, especially those occurring at third party refineries, are first communicated to supply staff through notifications of pipeline schedule changes. Because the schedulers making the log file entries had minimal time to investigate root causes it was suspected some portion of Refinery Supply Issues were included in the Pipeline Performance category.

In any case this type of information is not intended to be precise and does not define solutions per se. Rather it is intended to prompt questions and suggests directions for follow-up studies. Why would Required Terminal Maintenance cause so many truck loads to be delayed? Isn’t most of that work scheduled in advance? Can communications be improved to better inform truck dispatchers when it will be OK to load? With such a large percentage of issues cropping up in the pipeline’s span of control, would it help to hire some scheduling personnel that have pipeline experience? At any given terminal the profile of loading issues will be different; sometimes a revelation, sometimes restating what is already known. Either way the characteristics of the supply history are all in plain view for consideration at the front end. That sets the stage to bring the right stakeholders together and to get to work.