A Lesson Learned in Project Management:
(The error factor in a small difference between two large numbers)
Editor's Note: This Musing is a record of an actual project experience.
However, for obvious reasons, specific names of companies or individuals have
been carefully omitted. You can find details of the original project - Project
Downsizing - listed in the Papers section of this web site.
Leaving aside the snob effect of high-end goods like handbags and luxury cars, there is a general belief that "You get what you pay for". But that is not necessarily true. More to the point is that "You do not get what you do not pay for"! Worse yet: "You often have to pay for what you don't get anyway." The following example seems to be a case in point.
As a part of our Project Downsizing, we chose to send some of our furniture and belongings to our daughter in Toronto. To give you some idea of the challenge, the distance from Vancouver to Toronto on the Trans Canada Highway is around 4,500 kms. (That's 2,800 miles, although only 2,600 miles if routed through the USA). The minimum charge for a shipment of that kind is for a load of 500 lbs. To see where we stood regarding weight, we carefully drafted a waybill - see sample in Figure 1.
Figure 1: Sample of shipping list
After finding a recommended moving company, the normal process is to phone up, describe the goods to be shipped and the Customer Representative (CR) will estimate the weight and other factors and give you an estimated price over the phone. However, the caveat is that you will be charged on the actual weight shipped. This weight is determined by weighing the van on a government certified transportation scales before and after loading your goods.
Given the uncertainty surrounding the description of the goods by most customers and their propensity to change their minds at the last minute, this approach seems to be quite reasonable. But not entirely satisfied, we set about finalizing our list and actually weighing all the articles. For this, we used a regular household bathroom scale and we opted to include the "Large Oak Table" shown in Figure 1.
Now we recognize that the accuracy of a household bathroom scale can hardly compete with the unchallengeable accuracy of a government weigh scale. Still, we know through careful comparison with our doctor's mechanical scale that ours is not more than a couple of pounds out, or less than 2%. More to the point, however, imagine trying to weigh things like a large (and heavy) oak table, an oak roll top desk, or a sideboard.
Actually, it is not that difficult. What you do is weigh one end and double it. The trick is to get a plank of wood under a pair of feet of the given piece of furniture, and get the bathroom scale under the middle of that plank. You then just double the observed reading. There is a caution, however. Such scales do not measure accurately resting on a carpeted surface. Therefore, in this case, the bathroom scale itself must be resting on a flat surface such as a thick piece of plywood.
In any event, when we relayed our shipment request to the moving company, their CR's estimate of weight was very close to our own. All seemed satisfactory.
On the appointed day, we were a trifle nonplussed with the size of the moving van that showed up - see Figure 2.
Figure 2: The moving van
(No – This image has definitely NOT been stretched!)
More to the point, we were horrified when the bill came in - for a recorded weight not of around the expected 660 lbs. but for 1020 lbs. That is a discrepancy of 360 lbs. or over 50%. The obvious question, therefore, is what was it that was actually weighed under the loaded condition. Perhaps it included someone else's goods as well?
We called the moving company's president and provided him with the details and he promised to look into the matter. A few days later he called back to say that he personally had reviewed "all the paper work" and all seemed to be in order based on the details, and all weights appeared to be in order also. Nevertheless, the discrepancy did seem strange and he offered a refund based on "splitting the difference". There appeared to be no merit in pursuing the matter further so we accepted the offer.
But what is really at stake here, is the reputation of the moving company in particular and the integrity of the transportation system in general. How could such a discrepancy in weight be accounted for? There are a couple of theories.
The first and most obvious is simply that we are relying here on the difference
between two large numbers, namely the weights of the unloaded and loaded moving
van. (Take a look again at Figure 2.) Therefore, slight
differences in the calibration of two different government weigh scales could
produce a relatively large difference in the required weight.
The second possibility is more nefarious. That the migration of people, legal or otherwise, through Vancouver as a gateway to and from the Far East is a given. Sooner or later, some of these people landing in Vancouver will want to move out across Canada. What better way than to hitch a ride on a long distance transport perhaps acting as a swamper? If the trip is a clandestine one, and in hiding they happen to be included in the weight of the van, a couple of people could account for the weight discrepancy found in our shipment.
Unlikely? Certainly, but not impossible. In any case, the lesson to be learned here is: "Watch your weight" - literally!
1. If the large item to be weighed is not symmetrical in its weight distribution, then weigh both ends as described and add the two readings together.
2. A "swamper" is a truck operator's assistant who performs a variety of tasks as a helper under supervision of the operator but does not drive. A term especially used in the oil and gas industry.