Advanced search
Start date

Development of a software as a service to optimize the allocation of ceramic pieces into heat treatment furnaces


The production of high strength ceramic pieces has many phases. Initially, different types of ceramic powder are mixed and then compacted to define the shape of each piece. Next, the residues from the compaction process are removed from the pieces. After this phase the heat treatment is performed, where the pieces have to stay a period in high temperatures (burn) to gain rigidity and quality. Finally, a rectification is performed to finish the product. The heat treatment is a very relevant phase. In this process, the pieces are allocated in a furnace to burn, which can exceed 50 hours at temperatures of up to 1640 degrees Celsius. These furnaces are heated using electricity or liquefied petroleum gas and lead to a high use of resources by the industry. Furthermore, these furnaces are located in places that can reach high temperatures, generating discomfort to the employees. The selection of pieces to compose each heat treatment furnace as well as the accurate allocation of these pieces in the furnaces are important to reduce the burn costs. Predict how the pieces will be arranged, avoiding unbalances in the furnace is also important because it speeds up the furnace assembly process. If the furnace load moves during the burning process, some pieces can be damaged and then be discarded. To the best of our knowledge, there is no patent or reference to this problem in the literature. In this context, the objective is to propose an optimization software, that will be available as a "software as a service" (SaaS), to solve this problem, generating the best allocation of the pieces in the furnaces, maintaining the stability and reducing the number of furnaces required to perform the heat treatment in the ceramic pieces. By proposing a cloud optimization service, the industry will not need high performance computers or computers with specific configurations to solve the problem. The software should be intuitive, not requiring high technical knowledge to be used and it should be efficient in terms of time necessary to solve to the problem, giving more agility to the process in the industry. Also, the solutions should be easily interpreted, speeding up the piece selection and their allocation in the furnace by the employees which work in a place with high temperatures. To achieve these objectives, combinatorial optimization methods, such as mathematical models and heuristics will be used with web languages. The expected result is a SaaS application to assist the industries to plan the load of furnaces, with equal or greater quality solutions compared to the ones used in the industry, reducing the number of furnaces needed to treat all the pieces, leading to a direct impact on the amount resources used and the profit of the industry. Furthermore, the solution should be easy to interpret, reducing the time that the employee spends allocating the pieces, activity that is performed in a place of high temperatures. Also, it is expected to reduce the time needed for the selection and allocation of the pieces in the furnaces by the employees. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items