Nesting Algorithm Differences You Need to Know

5 Generations of Nesting Software

5 Generations of Nesting Software

When researching nesting software, it is very common for project managers to see all nesting software – even dynamic nesting – as the same.  However, the nesting software marketplace reality is very different.

As you might expect with all software, nesting software has evolved tremendously over the last thirty years.  What you need to know is that it has gone through five generations of evolution, and all five generations are still on the market today.  What you need to know is how to identify each generation, what each generation does and doesn’t do for you, and how each would solve your nesting needs.  It is the only way to make an informed, wise purchasing decision.

The generations are distinguished by the approach to nesting – how the algorithm addresses each part, optimizes for efficencies, and ultimately creates the nest.

First Generation - Rectangular Nesting

What is it? Rectangular Nesting “draws” a rectangle around the part at the largest height and width. It treats the part geometry as the rectangle, not the real shape of the part when placing the part on a nest.

Advantages. Rectangular Nesting is satisfactory if and when your parts are primarily rectangular in shape.

Challenges. This process does not consider arcs, holes, or other non-rectangular variations in the part when nesting. Similarly, rectangular nesting does not create the opportunity for interlocking of parts. A common example of interlocking parts are two L-shaped parts, one rotated 180 degrees, locking together like puzzle pieces. Also, holes are not filled with standard rectangular nesting software.

Second Generation - First Fit Nesting

What is it? First Fit Nesting Heuristics (algorithms) create an ordered list of parts. Most often the list is ordered from the largest part to the smallest part. The First Fit Nesting heuristic places the largest part in the list on the nest first, then the next largest and so forth. If the second largest part doesn’t fit, the software moves down the list to the first part that will fit; hence the name “First Fit.” Additionally, when considering the part for placement the nesting tool chooses from several pre-set rotation options (90°, 180°, 270°) to find the best fit. Best fit is defined as rotation that brings the center of gravity closest to the lower left corner (or other specified datum point).

Advantages. The First Fit Nesting approach is more automated than manual nesting and can be less time consuming.

Challenges. There are several limitations to the First Fit Nesting tactic. It is impossible to create a single list that will reflect all of the demands on the production schedule, i.e. due dates, hot parts, while maintaining a largest to smallest part order. Nesting mathematics is very complex. Since 50 parts can be nested in more than 10100 alternative ways, this single list is only one of the many possible nests and is extremely unlikely to be close to the optimal solution.

Another challenge is the limited number of part rotation attempted. As an example, assume a rotation setting of 10 degree increments – 10, 20, 30, 40, etc. – and can only be rotated in those increments. If a part must be rotated 92° to fit, the part would be rejected as not fitting in the space available. If the software is given a large number of rotations, the time to nest the parts can become impractical. In short first fit heuristics are blind and are not able to consider multiple requirements simultaneously.

Despite these limitations, the first fit heuristic is used widely by a number of nest software suppliers. The reason that this second generation heuristic is used so much is that it is easy to code and easy to understand. Third, fourth and fifth generation nesting technology is very complex.  Many second generation software suppliers offer multiple variations of the first fit method, which they consider as different nest algorithms. The user can run the parts through one algorithm, see the results.  Then the user can run it through the second algorithm, see the results, and compare it to the first algorithm, and so on.  It can be pretty laborious.

 

 

Third Generation - Half Shape (True Shape) Nesting

What is it? Half Shape Nesting identifies a portion of the actual shape of the part. It puts the shape in the lower left corner of the space available and identifies the minimum “X” and minimum “Y” coordinates where the next part can be placed. Often Half Shape nesting is called True Shape nesting because it uses the actual part boundary as it places the part. However, only half of the part shape is considered. Only the left side and bottom of the part is examined to determine how well it fits with adjacent parts. The top and right side are ignored until another part is placed next to it. In Half Shape Nesting algorithms, the parts already placed on the nest remain stationary and only the newly inserted part is considered for placement and rotation.

Advantages. Half Shape or True Shape Nesting is a more real-world approach than Rectangular Nesting, because it takes into consideration half of the actual shape of the part during placement. When the part shape is used, the nesting tool can find greater material savings advantages by rotating the part. It also opens up the possibility of greater nesting efficiency.

Challenges. Half Shape or True Shape Nesting comes up short in its ability to make evaluations about the full shape of the part. Some questions it fails to answer include: Is the next part the best part to select for this location? What is the best orientation for a group of parts? Half of the part may fit well with existing parts on the nest at some odd orientation, but that may cause subsequent parts to cascade into a random inefficient

Fourth Generation - Multi-Dimensional Combinatorial Nesting

What is it? Multi-Dimensional Combinatorial Nesting is another automatic nesting technique. The software uses mathematical fathoming to eliminate alternatives that do not need to be considered. See the Flash presentation for a full explanation of fathoming. The nesting software automatically and intelligently considers only those part combinations (nests) that take into consideration machine efficiency, schedule demand, order completion, material efficiency and many more real world requirements. Part layout solutions that are outside of the optimal solution set are simply not considered. In this approach, the production priorities are part of the expert knowledge base in the nesting software enabling it to make intelligent decisions. Due dates, hot parts, machine efficiency, material cost, part attributes and more are evaluated then optimized into a nest or series of nests that the optimal solution to the user’s requirements.

Advantages. This method significantly reduces programming time and retains the best possible results for all considered factors – schedule, material, order completion, etc. Benchmarks show 8% to 16% higher material utilization over other methods.

Challenges. While the technology can be simplified and used in any environment, the expert system technology can best be leveraged by fully integrating the system with other manufacturing systems such as ERP/MRP, CAD and other common manufacturing tools. Training and a good support system is necessary to gain the maximum benefit from the technology.

Fifth Generation - Vision Emulation

What is it? Vision Emulation is a feature of fifth generation nesting technology. Vision Emulation Nesting “sees” the actual full shape of the part and makes logical conclusions about it, just as a human looking at the part would. The process is modeled after human vision and decision making.

Vision Emulation looks at the full shape of the part and the space available on the nest, then determines if there is an optimal fit.

Vision Emulation also evaluates the part to determine if and how much rotation is needed to provide an optimal fit. The actual part shape and the shape of adjacent parts is used to determine the optimal orientation. Multiple parts may be viewed on one time. A part could be rotated 123.456 degrees to achieve an optimal fit. This process eliminates the time consuming trial and error process of rotating the part in hundreds of small increments to check for fit. To understand the advancement that Vision Emulation provides, imagine putting a puzzle together in the dark. Without the ability to see the puzzle piece, you would have to try many orientations to determine if the part fits. Vision Emulation is like turning on the lights.

Advantages. Visual Emulation can automatically find occasions to reduce material waste by seeing parts to nest in the appropriate voids. It naturally reduces the nesting time by allowing the optimal placement to be seen – unlike previous generation nesting – by only trying reasonable shaped parts and orientations that are optimal.

Questions?

Contact Optimation for a discussion of the best nesting algorithm for you.

Notice: This work is licensed under a BY-NC-SA. Permalink: Nesting Algorithm Differences You Need to Know

Leave a Reply

Your email address will not be published. Required fields are marked *

*

240,415 Spam Comments Blocked so far by Spam Free Wordpress

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>