We are all familiar with the famous disclaimer: “Past performance is not necessarily indicative of future results”.  In our world, that of HISTORICAL K-factor deliveries, past performance is heavily relied upon – mostly successfully.  Yet, the pandemic that swept our nation over the past 7 months will put that success into question in a very real and immediate way.

Work from home, distance learning, etc., has, without warning, from last winter to the upcoming winter, made much of the historical analysis simply incorrectThere will be more consumption per HDD for homes that have more people working and learning in them.  It is a simple as that.  After each delivery, we can update the Ks and try to catch up.  However, the historical calculations are built (correctly so) to NOT overreact when one delivery is not in line with expectations.  How about two deliveries?  How about three?  The problem is going to be that not only do homes not consume fuel at a pace that it directly linked to a “base level of HDD’s per gallon”, but now in many cases that base level will have changed.  Put another way, picture if all your customers had perfect Ks for 10 years and then 25% of them suddenly added another 750 sq. ft. onto their homes – and none of them told you.  How long until you are caught up?  How many runouts would you have?

As an oil or propane distributor selling to automatic delivery customers, you face two absolute extremes relative to the core of your business, delivering fuel.

And so, every dealer chooses something in between.  Dealers try to balance the number of deliveries made such that there is a low risk of runouts, while not making too many (smaller) deliveries that drive up costs.  A quite common approach is to accept that SOME runouts will occur, and hope that the number is “acceptably” low.  All of this depends on some methodology to predict tank levels and consumption.

Regardless of how you schedule your deliveries, the biggest variable in that planning process is “how many gallons are in the tank on any given day”.  For tanks without remote monitors, the daily tank level is provided by your Back-Office System (BOS), and uses a combination of factors including factual, historical, and predictive.  Each BOS has a proprietary approach toward calculating consumption (from the last delivery until today), and projecting consumption (from today until a point in the future when you need to make a delivery).  While acknowledging that there are small differences from one BOS to another, they all take a similar approach and do a relatively good job.  If you know the tank size, the Usable Size (entered by the dealer to determine how many gallons should be “used” for consumption), the Optimal Delivery (the targeted delivery size), the Reserve (the flip side of the optimal delivery, considering the usable size), Fuel Use (whether the home uses the fuel for heating only or for other uses: hot water, clothes drier, generator, etc.), Historical HDDs, and Historical Consumption (defined by a certain number of deliveries over time), you have the elements needed:  facts, history and a calculated guide (K-factor).

For decades, the math has not changed, the calculations have not changed.  While the BOS is doing what it is supposed to, there are still some runouts, and there are still some unusually small deliveries.  Is the BOS to blame?  Most likely, not!   The difference between anticipated delivery size and actual delivery size is not the fault of the BOS and is not the fault of incorrect mathematical calculations.  It is simply because homes do not consume fuel in a way that is perfectly correlated to the weather.  When you deliver to a customer 6 times a year, you have only those six opportunities to gauge the relationship between weather and consumption.  None of this is new and it is the reason that:

(a) planned deliveries are smaller than usable tank sizes, and

(b) more tanks than ever have remote monitors

Does that mean that all tanks must have monitors and that K-factor based deliveries are an awful way to approach delivery logistics?  No, not at all.  If you have enough history you probably can figure out which homes are “predictable enough” and which ones are not (or at least that used to be the case).

In a world of rising expenses, access to data, and a desire for operational efficiency, it is doubtful that you simply want to (or can) just make more smaller deliveries to all of your customers so that you can avoid the impact of “that unidentifiable 25%” that will be burning more fuel.

Possible Solutions:

  1. Remote Monitors can help update Ks by reporting consumption 365 days a year, not 6 times per year. If you do not know what is in the tank and you do not know how consumption has changed until you are making a delivery, how will you avoid runouts or short deliveries?  This winter will represent the biggest challenge that we have ever had in that regard.  If you have monitors, you need to get them installed.  If you do not have monitors, you need to investigate them quickly.


2. Your customers are aware there is a pandemic. Perhaps you should be reaching out to them.  Explain to them that one of your main objectives is to ensure they do not run out of fuel.  Let them know (unless they already have a remote monitor) that the relationship between the weather and their homes’ consumption may have changed if they have more people at home during the day than they used to.  Poll them.  Ask them if they think they will be burning more fuel.  If they say yes, go into your BOS and lower their K.  The BOS will eventually catch up and balance that change, but if you KNEW who would be consuming more fuel, you would behave differently and not take on as much risk.


3. Lacking either of the two choices above, it may pay for you to just lower everyone’s Ks by a bit. This may seem to be addressing just a quarter of your customers, while being overly cautious about the other 75%; that is exactly what it is.  However, if Ks were no longer indicative of consumption, wouldn’t you be more conservative?


It is best to KNOW how much is in a tank.  If not, you need to manage your delivery planning based upon rules that have changed.  If you are still using the same delivery planning that you have always used, or you are using the more modern and optimized ADEPT approach, neither will create foolproof results with the new rules.  It is not anyone’s fault – not yours, not the BOS, not your dispatcher, not ADEPT.  You just need to learn to play with the new rules of the game.


Written by: Philip Baratz, president of Angus Energy