This article is part of a series of articles. Previous parts of this article series can be found here:
Cockroach Swarm Optimization
No matter where you go on this planet (except for Antarctica), cockroaches are everywhere. These highly resilient buggers have been around for a while. Fossils discovered by archaeologists prove that cockroaches have been around for more than 300 million years. They are one of the oldest groups of insects on the earth and one of the most common. Scientists estimate that there are around 4,600 species of cockroaches worldwide. Some of the traits that make them survivors are also traits that inspire the cockroach swarm optimization algorithms, as shown in Figure 1.
Figure 1: Overview of cockroach swarm optimization
As you may pick as a consistent theme in many swarm algorithms, these algorithms are heavily centered around foraging i.e, searching for food or prey. The cockroach swarm algorithm is no different. However, multiple cockroach biological behaviors are imitated in cockroach swarm optimization. These include chase-swarming, dispersing (when you turn on the lights and these buggers run for cover), searching for food, relocating nests, and being ruthless. If your nerd radars need more inputs on all swarm algorithms, I suggested a book in the first part that covers all the swarm algorithms, with their mathematical formulations and pseudo codes.
Applications of Cockroach Swarm Optimization
This algorithm seems to be a good candidate for VRP, as far as applications in supply chain and operations go. Analyzing some capitative studies fairs better than the ant colony algorithm when solving the Traveling Salesman Problem (TSP). You can refer to these research papers for specific approaches leveraged to apply this algorithm for solving TSP.
Cockroach Swarm Optimization Algorithm for Travel Planning
Cockroach Swarm Planning for TSP
A New Way to Solve Traveling Salesman Problem
Remember the bat swarm optimization article where I suggested this algorithm looks like a good fit for bot mobility planning? Looks like the cockroach swarm optimization algorithm is a much better fit (see the research paper link below). Like I postulated in that article, this algorithm can be leveraged in scenarios where numerous bots are on the floor to chart the path of these bots optimally. Back in 2020, I wrote an article suggesting that perfecting swarm algorithms and coupling it with an aligned infrastructure is the key to realizing the dream of scaled drone delivery. The same postulate applies to bot-managed, fully autonomous eCommerce warehouses (which are not on the near-term horizon).
A New Cockroach Swarm Optimization for Motion Planning of Mobile Robot
In my perspective, the benefits of improving the solutions to TSP are marginal, when it comes to vehicle fleet routing in the real world. However, since warehouses will increasingly become more and more automated, the application in optimal floor mobility planning can be differentiating if one organization develops it much before others.
Like some other swarm algorithms, this algorithm can do a better job than other widely used classification algorithms. It has been tested in mage classification scenarios, combining it with Machine Learning (ML) algorithms like Support Vector Machine (SVM), like this example. This approach can be extrapolated into defect detections (quality management) in smart manufacturing environment.
For the more nerdy ones, this research papers shares an application for density based clustering.
Metaheuristic Based Clustering with Deep Learning Model for Big Data Classification
However, it is the image detection category that seems to be pretty powerful. One of the powerful applications is in the Bioimaging domain, where this algorithm, in combination with other approaches (primarily NNs), can help improve the detection of diseases. Here are numerous examples that, though not supply chain-related, definitely show the potential that these algorithms can impact the well-being of humankind. Who thought that cockroaches could be helpful!
Multiple Eye Disease Detection using Hybrid Adaptive Mutation Swarm Optimization and RNN
A Deep Learning Based Cockroach Swarm Optimization Approach for Segmenting Brain MRI Images
An Optimal Thresholds for Segmenting Medical Images Using Improved Swarm Algorithm
In the subsequent article in this article series, we will explore crow search algorithm. The article will be published on 12/6. For now, give a little respect to the tiny cockroach next time you spot one.
Designed Analytics Blog 