Published in the Journal of Expert Systems with Applications, the CSM-funded study investigated the use of two common evolutionary algorithms – Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) – to improve production efficiency in bakeries.
PSO is an adaptation of the movement and behavior of bird flocks or fish schools searching for a food source and ACO follows the mechanisms that help ants to find the shortest routes between a food source and their formicary.
The researchers found that use of a carefully designed production model based on the scheduling theory in combination with PSO and ACO algorithms improved operating efficiency.
“Since ACO and PSO both provide easy implementation, easy modification and the ability to solve complex scheduling optimization tasks in reasonable computational time, they are promising methods to solve bakery scheduling tasks,” they said.
The production model and optimization were programmed and performed with MATLAB – a technical computing tool from MathWorks that enables users to analyse data, develop algorithms and create models and applications.
The 'swarm' in a PSO algorithm consists of particles or possible solutions - which are the product sequences in a bakery production plan - which reflect individual swarm members in nature.
"During the iterations of the algorithm the particles are 'flying' through the search space. Due to the frequent update and comparison of the best sequence so far and each particle’s current and so far best value of the cost function, the particles move to the optimal solution of the given optimization problem," the researchers explained.
Optimizing makespan and idle time of machines
The researchers investigated how implementation could ultimately cut costs for bakers. They looked into optimizing makespan – the required time to complete the given production goal – and idle time of machines – how long a machine or stage is ready and free to process a job.
Both algorithms can solve each optimization problem in less than 15 minutes, they said, providing a procedure that can be employed shortly before the start of production.
“…Even the worst case results of both algorithms for both cost functions yielded significant benefits compared to the results given by the initial product sequence,” they said.
Clever planning needed for time-sensitive production
Taking the German bakery sector as an example, the researchers said that because bakery production planning is often based on the practical experience of the responsible employees in the bakeries, rather than on mathematical methods, it leads to “sub-optimal performance”.
Clever planning is needed for bakery production, they said, particularly given how time-sensitive the processing is.
“The production of bakery goods is strictly time sensitive due to the complex biochemical processes during dough fermentation, which leads to special requirements for production planning and scheduling.”
The PSO and ACO algorithms can work to improve the efficiency of production because they enable bakers to calculate an optimal plan and determine the exact time schedule and capacities of devices needed to reach the production goal.
The research data was collected from a bakery in Frankfurt.
Source: Journal of Expert Systems with Applications
Published December 2013, Volume 40, Issue 17, pages 6837–6847
"A case study on using evolutionary algorithms to optimize bakery production planning"
Authors: FT. Hecker, WB. Hussein, O. Paquet-Durand, MA. Hussein and T. Becker