Mathematical sequencing created using algorithms will improve the efficiency of any bakery, a scientist says.
Research recently published in the Journal of Expert Systems with Applications said that use of animal-inspired algorithms could improve production efficiency and cut costs in bakeries.
Speaking to BakeryandSnacks.com, lead researcher of the study Florian Hecker said that humans cannot develop bakeries anywhere near as efficient as those created using algorithms.
“A human is just not able to calculate all possibilities of product sequences and thus is not able to determine the sequence providing optimal performance,” Hecker said.
“The empirically developed sequencing is of course not completely wrong, but sub-optimal.”
‘Enhanced’ production planning
In his study, Hecker and his colleagues used the German bakery sector as a case study and found that many bakeries designed production plans on the practical experience of the responsible employees, rather than on mathematical methods.
They said that for bakery in particular, “clever planning” was needed given the time-sensitive nature of the processing.
“The use of numerical methods just provides new possibilities for creating an enhanced production planning, because they are able to find an optimal or near optimal solution out of billions of possible sequences,” Hecker told this site.
A money-saving investment
Hecker’s research used common animal-inspired algorithms Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO), implemented using a technical computing tool from MathWorks.
He said that while the return on investment was hard to predict, implementation of systems like this would certainly benefit users and save money.
“The presented tool could enhance the production planning and thus the efficiency and performance of a company,” he said.
Hecker said the benefits of mathematical systems would not be dependent on the size of a bakery, but rather the usability – in terms of how complex the production is and how many different products are being manufactured.