A structured model to manage a large number of transactions

The aim of this paper is to design a model for managing the daily transactions for one of the most important Italian banking groups. This model is expressed using an object-oriented language called BLOOMS proposed by the authors, which fully follows the core concepts of the Structured Modeling as discussed by Geoffrion in 1989. The transactions, which can be partitioned by type, have to be satisfied within certain time constraints. This makes the storage of data not an easy problem to solve. These data in fact are stored in four levels of memory, which have a different transfer rate and, of course, a different cost function. The large number of transactions to be dealt with oblige us to consider their distribution within a specified time period. The goal of the model is to maximize the throughput of the system coupled with the minimization of the storage cost under time, space and budget constraints. The outline of the paper is as follows. In the first section the entire problem is described; its mathematical formulation is given in Section 2, while in Section 3 it is instanced following the scheme given by BLOOMS. A numerical example with its solution is given in Section 4. Conclusions end the paper.