When evaluating the performance of a scanner system, several methods and criteria can be used. One of the most important, but also not so well known indicator, is the Value yield.
The value of a board is usually calculated from price (in €/£/SEK/$/..) per m³, or directly as an item price for each product (dimension / length / grade). When calculating the board value, the price of waste products (chip, sawdust) must also be considered. The goal of the sorting process is to achieve as high value of the produced items as possible. Is it better to make a short high grade solution, or a longer lower grade solution? The value optimization will make the decision.
Value yield is a number given in %. It describes the actual value you get from your production, compared to the theoretical maximum value. In a perfect world the value yield would be 100%; in real life it will be lower. The goal of the sorting process is to get as close to 100% as possible. The importance of the value yield as an indicator of the process performance, can easily be understood. The higher the value yield is, the higher the profit for the sawmill will be.
There are many reasons why the actual value yield will not be 100%. One reason, which usually doesn’t come first into mind, is that there may be disagreements between people what 100% actually is! One person says that a board is grade A, another grade B, or one person measures a knot to be 25 mm, the other 2 mm bigger. Who is right? Everyone, who has made precision measurements of board defects knows how difficult and arbitrary that can be.
When evaluating the value yield of a sorting process (manual or automatic), the process would typically be as follows:
- Boards are selected for the test
- The boards are numbered and manually evaluated in order to establish the 100% level
- The boards are run through the system
- All decisions by the scanner or the operator are registered (grades, lengths, trims, etc.)
- The final trimmed boards will then be re-evaluated, and their actual values will be decided.
The value yield can now be calculated. It will be the sum of the values of the produced boards (ProdSum), divided by the sum of the maximum values for each board (MaxSum):
Many studies have been done about what levels of yield are achieved in manual sorting, and they indicate that they typically would be 80-85%. With automatic sorting the goal can be set much higher, reaching levels of 96-99%. The BoardMaster system is always maximizing the value yield, and therefore also the profits of the sawmill.
Selecting a supplier for automatic grading
When selecting the supplier for the sawmill grading line there are a few things you should take into consideration.
Where and how?
The first thing is to determine where and how the quality grading is done: at the green mill or the dry mill or at both locations? How about the layout? At the sawmills the scanners are normally installed transversally, because that gives the highest efficiency and is generally most suitable for the layout. What is the maximum speed the automatic system can manage? Old lines can typically manage about 80 - 120 lugs / minute, but many new lines are designed for a speed about 200 lugs / minute. Is the proposed grading system capable for higher speeds?
Experience of the supplier candidate?
One key point in suppliers is their experience:
- how many systems have been delivered?
- which process areas are covered: edger, green grading, dry grading, planer?
- which applications are covered: rough or planed boards, strength grading, wood species?
Sawmill business is quite global, and mills typically deliver their products to many different markets. Does the supplier have the ability to assist with implementing various grading rules (Scandinavian, British, American, Russian, Japanese)? Is the user interface and documentation available in your local language?
The grading scanner is connected to the line controls. Does the supplier have experience in working together with your local automation and machinery supplier?
The rapidly growing machine vision technology includes a lot of technical details, which might be confusing. Looking at perfect defect detection can cause inability to select the supplier. Does the scanner analyze the whole board? How well is the geometry of the board analyzed? Is the pith location detected? Is inner and outer surface determined? Are all “your” defects analyzed with high accuracy?
The key issue is to look at the grading results. The biggest economical benefit from an automatic grading system comes from the value yield which should be at least 96%. Does the supplier have recorded results?