M1 – ISCE/BI
Operational Decision Making Systems
Workgroup
Duration: 10hours
Group: keep the same groups as on the first workgroup
Problem description
Numerically controlled machines have a tool magazine that stores all the tools needed to run a range. A tool changer is used to interchange the tool in place on the machine with a tool from the store. There are different types of stores and tool changers. For this study, we will consider two-way circular stores with a tool changer that performs tool permutations.
Optimization of the tools position
in the loader of a numerical control machine
This operation implies that the change of tools can be done partially in hidden time. Indeed, during the period during which the tool in place is used, the store can turn to bring the next tool. It should be noted that a tool does not necessarily remain in the place it initially had.
The arrival of high-speed machining machines and the appearance of long machining ranges (in terms of the number of operations) and involving brief operations no longer make it possible to perform tool changes in hidden time. Therefore, the time taken by the machine to change the tools becomes a non-productive and non-negligible time compared to the total machining time.
The position of the tools in the store influences the changeover time of the tools. One can therefore wonder what is the “good” position of the tools in the store that reduces tool change times.
Operational Decision Making Systems – 2018- 1
Objective of the Project
This project consists in creating a tool allowing, for a given range, to optimize the position of the tools in the store of a machine. This tool will be based on Simulated Annealing (SA) method.
Set of test
Two sets of test are available on Moodle.
Work to do
Formalization of the problem
Determine input data and decision variables.
Calculate the total running time of a range for a given range.
Since the sum of the machining times is constant, determine the objective function of this
problem.
Looking for an optimal solution – Implementation of the Simulated Annealing method
Provide a coding that allows representing the problem.
From the coding of a solution, propose a method to evaluate the objective function.
Propose a neighborhood method that ensures the connectivity of the search space solutions.
Propose a method allowing to obtain an initial solution.
Program in language C the Simulated Annealing method that corresponds to this problem and adjust
the parameters from sets of test.
Apply this method to other sets of test and check the setting of the parameters.
For winter evenings…
Does duplicating a tool save time?
Operational Decision Making Systems – 2018- 2