1,721,454 research outputs found

    Optimisation of a simulated-annealing-based heuristic for single row machine layout problem using genetic algorithm

    No full text
    We discuss a procedure to determine the optimal set of parameters relevant to heuristics based on the Simulated Annealing technique, an algorithm which is widely applied to combinatorial problems in the field of manufacturing systems. We consider the search for the best set as a second optimisation problem that we solve by a Genetic Algorithm. The performance of our approach is tested in the particular case of backtracking minimisation in a single row machine layout problem for flexible manufacturing systems

    Material handling devices dimensioning in multi-stage assembly line

    No full text
    This paper deals with the problem of dimensioning the number of material handling devices (MHDs) needed to supply a given (single-product) multi-stage assembly line, in order to obtain a pre-determinate risk level of incomplete items produced. The assembly line is managed in accordance with the criteria imposed by the just-in-time philosophy. An iterative analytical model based on the queuing theory is developed. This model permits not only the dimensioning of the number of MHDs but also the resizing of the component containers capacities, in order to reduce the variance of the interarrival time of supply requests from the assembly stations. A simulation study confirms the precision of the results obtained. A brief discussion about the mixed-model assembly lines is also presented

    Designing a belt conveyor controller in bottling plant using fuzzy logic and genetic algorithms

    No full text
    This paper presents an industrial case study relevant to a fuzzy logic controller designed via a properly developed genetic algorithm. We consider an example of a fuzzy logic-based industrial process-controller. In particular, we deal with the problem of controlling the speed of a belt conveyor for glass containers in a bottling plant. The primary objective of the controller is to guarantee the continuous feed to the filling station, in the presence of frequent gaps between bottles. The secondary objective is to reduce the impact speed between arriving bottles and those standing in the queue, in order to reduce the plant noise. High-performance parameters of the fuzzy controller are found by a properly developed genetic algorithm. The results provided by Monte Carlo simulations demonstrate that, with such controllers, it is possible to achieve both the objectives mentioned above

    A study of non-linear autoregressive processes. Differential theory

    No full text
    It is shown how the general analytical expression of an autoregression, whose stationary solution has an arbitrary given form, can be obtained. The central-limit theorem is used to state a correspondence between autoregression and relevant diffusion equation which not only permits to give analytical form to the stationary distribution of a given autoregression, but also to obtain appropriate series expansions of its fundamental solution and the exact relaxation constants. The interest of the analysis in the context of Monte Carlo simulations of relaxation and steady-state processes is discussed. The procedure is illustrated by two examples of interest in the field of ionized gases

    A dynamic inspection model for wearing production systems

    No full text
    The aim of this paper is to correlate the deterioration process of a production system with the items control process and, as a consequence, with the equipment inspection policy. The objective is to consider how the wear of the manufacturing system influences the values of the control process parameters and how these parameters have to be changed in accordance with the need to maintain optimal (working) conditions. Adopting a Weibull failure-time distribution, it is shown how these parameters must be changed (at different times t) in order to maintain optimal working conditions. To test the inspection model, a weighted cost function dependent on the time loss for the production of defective units and for the production shut-down (in order to make an inspection to control the state of the system) is developed

    Heuristics for single row layout problems in flexible manufacturing systems

    No full text
    A series of algorithms relevant to one of the commonly used layout schemes in flexible manufacturing systems, that is, the single-row layout, is presented. The algorithms constitute a modified version of a known technique developed for the n/m/F/Fmax sequencing problem. As a result, heuristics are obtained which are easy to implement and which generate good solutions in acceptable response times and outperform other algorithms developed to the same end. An iterative method based on a neighbourhood search is also developed to evaluate the performances in absolute terms

    On the variance of stock price distributions

    No full text
    A stochastic approach is used to describe the temporal behaviour of stock prices. It is shown that standard techniques are unable to give the correct variance of the stock price distribution if the trend of variation is price dependent

    A note on stock price volatility

    No full text
    A new model of stock price volatility is discussed which avoids difficulties of other approaches. It is shown how the phenomenon can find a quite natural (possibly partial) interpretation in the correlation between trend and price. The contribution to volatility of the mentioned correlation can be calculated with various techniques one of which is treated here in detail

    A study of non-linear regressive processes. Integral theory

    No full text
    Autoregressions with constant parameters are special cases of a general iterative process which, in this paper, is given explicit form. No restriction is imposed to the process of variation of the random variable which is permitted to suffer both continuous variations and/or discrete jumps. As a consequence, the general autoregression assumes the same form of the iterative solution of an appropriate integral (master) equation. The procedure which leads to this conclusion is simple but not so direct and requires a special expedient that in this paper is discussed in detail
    corecore