Site Map | Homepage Contact us | Search | Glossary | Accessibility | Subscribe

Tendertips

The Problem

Concept drawings

Concept drawings on the whiteboard
(click to enlarge)

camera

Working model of the camera
(click to enlarge)

camera

The camera in its box above the line
(click to enlarge)

camera

Looking up inside the camera box
(click to enlarge)

COP Brief development

When Japanese customers buy a 100 gram bundle of Tendertips asparagus, they have firm expectations:

  • The bundle will not weigh less than 100 grams
  • They can choose from bundles having 3,4,5,6, or 7 asparagus spears, according to their preference
  • Within the bundle, the spears will be of closely similar diameter. This is important because they must cook to the same extent in the same length of time
  • The spears will be straight or very slightly curved
  • The spears will be almost circular in cross section
  • The spears will all be the same length
  • Where the asparagus tips curve, the tips will all point to the centre of the bundle
  • The bundle will be held together by 1 green elastic band, and 1 piece of food-grade adhesive tape

Many of these requirements may be regarded as cosmetic, but they are nevertheless essential.

How are these criteria achieved?

Essentially, the machine mentioned in the previous section examines each spear in turn as it passes along a moving belt. The machine detects and rejects spears which are very bent, out-of-round, or have white patches.

The remaining (good) spears are trimmed to the same length, and an estimate of their size/weight is made. The weight determines which kind of bundle the spear will go in (3, 4, 5, 6, or 7 per bundle). All this is done at a rate of 12 spears per second, and uses a special video camera to capture an image of each spear which is analysed by software to detect any faults. If fault-free, the software estimates the weight and passes control to another program.

The second program controls the machine to gather similar spears into bundles, counts the number of spears in the bundle and makes sure the bundle weighs at least 100 grams. As each bundle is completed, it drops into a tray whence it is collected for further processing. The final stages, where the loose bundles are aligned, banded, taped and packed into boxes, are done by skilled people.

camera and line computer

camera and line computer

The camera and line are computer controlled from inside tha packhouse

Scales

Scales check the system's accuracy

COP Brief development

This process produces a perfectly acceptable product, but at considerable cost. Two main areas of potential savings were identified.

1. Elimination of remedial work on bundles produced by the machine

The existing machine could not determine the size, or estimate the weight, of individual spears to consistently close tolerances. Each bundle required examination and many required non-conforming spears to be replaced, to make the bundle visually pleasing. The improved bundle then needed to be re-weighed. In addition to the extra cost of the staff involved, this stage of the process became a bottleneck limiting the throughput of the plant to well below its theoretical maximum (the machine speed of 12 spears/second).

2. Overweight bundles

As a direct consequence of the machine’s inability to accurately estimate individual spear weights, the actual weight of each bundle often differed markedly from the nominal weight of 100 grams, and the standard deviation of the error was many grams. To ensure that no more than 1 bundle in 100 would weigh less than 100 grams, the machine had to be set to produce a bundle size of 100 grams plus three times the standard deviation of the error. Over a long period of time the machine will match this target on average. This means that the customer pays for 100 grams, but on average gets 110, 120 or even 130 grams!

The laxity of the machine’s control of bundle weights results directly in the company giving away lots of perfectly good asparagus, free.

It can be seen that both cost-saving opportunities can be achieved by a single means; improving the machine to make much better estimates of the size and weight of individual spears of asparagus. This, then, was the objective of the first stage of the project.