ITE Consult at The AnyLogic Conference 2019 #ALConf19

The AnyLogic’s seventh annual Conference took place at Austin, Texas (USA) the 17 th –
19 th of April. This year’s conference featured panels with teams from all over the world.

This event has been a great opportunity to connect with managers and engineers from
different places and see what is new in the field of simulation and optimization
solutions.

ITE Consult has been be part of the first day of the conference and has exposed a case
study where we introduce the innovations and technology for Supply Chain Solutions
in our company.

Wednesday April 17th, 2019
2:00 PM – Consumer Package Goods:
Detail Schedule Optimization

ITE Consult Team: Elisa Elena, Engineer; Gaston Fourcade,
Javier Cortes, Computer science

For more information please check our presentation at www.ite-consult.com

CASE STUDY

Optimization of a Production Detail Schedule
A Multi Phases Pet ’s Food Production Process.

The Challenge

According to a weekly changing demand of more than 140 SKU’s, the problem to solve
was how to produce the expected demand, optimizing each of the phases (extrusion,
drying, coating, kibbles storage and packaging) that were restrictive over the others,
and changing from fluid to discrete. Individual phases capacities were large enough to
make the user feel he could produce the input demand, but there was no way to
achieve this goal without the help of this simulator.

The Concept
This optimizer has been designed to be a trade-off solution, between the optimization
of the partial components.

The Solution
A simulation and optimization model built with Anylogic, were we combine discrete
and fluid behavior. The Fluid library has been extremely useful for the multi
components SKU production, as it allowed to combine different master formulas for a
certain finish good.
We also developed an agent that put together a combination of strategies, to allow
each phase to make their own decisions having a global view of what is happening in
the other process steps in order to make a better decision, before optimizing the
combination.
Due to the large number of possible combinations it was not feasible to develop a
standard optimization model, so the use of Anylogic agents, with the internal strategic
functions we have developed, allowed us to achieve the optimized Detail Schedule as
input for each step fulfilling demand when feasible and optimizing the use of
resources.
The use of Optquez, in the optimization experiment, was the final step to get the best
weekly detail schedule.
Other features we have used are the ability of Anylogic in choosing the excel file to be
used and compare runs features, the ability of connecting Anylogic with Python, for
previous and outputs reports and graphical interphase.

The Output
The output of this simulation/optimization is an optimized detail schedule for each of
the process phases.
As a result, we have been able to maximize the tons produced and minimize the waist
in the different stages, with big savings for the company.
Actually, the company is using the model, for their weekly detail scheduling, to be able
to fulfill their demand and avoid waist.