كاربرد روش‌شناسي رويه‌ي پاسخ (RSM) و روش رگرسيون فازي درتعيين ميزان بهينه‌ي عوامل موثر در فرايند مونتاژ كاسه‌چرخ خودرو

APPLICATION OF RESPONSE SURFACE METHODOLOGY AND FUZZY REGRESSION METHOD TO DETERMINE OPTIMUM AMOUNT OF EFFECTIVE FACTORS IN VEHICLE BRAKE DRUM ASSEMBLING PROCESS

با توجه به نقش حساس مجموعه اكسل، بالاخص كاسه‌چرخ خودرو، به‌دليل مرتبط‌بودن آن با ايمني سرنشينان، بررسي فرايند توليد و مونتاژ و نيز انجام آزمايش‌هاي كنترل كيفي حين اين مراحل از اهميت به‌سزايي برخوردار است. در اين نوشتار با توجه به اهميت بالاي سه عامل اصلي قطر توپي قسمت كاسه‌نمد، قطر داخلي كاسه‌نمد و گشتاور مهره قفلي به‌عنوان متغيرهاي مستقل، مي‌كوشيم ميزان گشتاور چرخشي كاسه‌چرخ خودرو را (تحت عنوان متغير پاسخ اول)، به‌كمك مباحث مطرح در طراحي آزمايش‌ها و روش‌شناسي رويه‌ي پاسخ و لقّي كاسه‌چرخ (تحت عنوان متغير پاسخ دوم) به‌كمك روش رگرسيون فازي با كاربرد برآورد‌كننده‌هاي كم‌ترين انحراف مطلق بهينه ‌كنيم. در نهايت جواب بهينه با استفاده از روش LP‌سنجي بررسي شده است.

Considering the crucial function of the axle assy, especially the vehicle brake drum, due to its relevance to the safety of the passengers, studying the production and assembly processes and conducting quality control experiments during these stages is of great importance. In this study, two approaches are used to improve the vehicle brake drum assembling process. The first approach is named; response surface methodology (RSM). Response surface methodology is the most popular optimization method used in recent years. There is so much work, based on the application of RSM in chemical and biochemical process. The effects of process parameters for the vehicle brake drum assembling process were exploited using the; design of experiment (DOE). In this work, experiments are performed by a standard RSM design, called a central composite design (CCD). With regard to the great significance of three main factors, namely; seal-oil spindle diameter, seal-oil internal diameter, and nut lock torque, as independent variables, the present research attempts to optimize the rotatory torque of the automobile brake drum (as the first response variable) obtaining assistance from discussions regarding the design of experiments and the response surface methodology. The second approach is named; fuzzy regression. There is a likelihood that the greater the values of independent variables, the wider the width of the estimated dependent variables. This causes a decrease in the accuracy of the fuzzy regression model constructed by the least squares method. In this paper, we use the least absolute deviation estimators to construct the fuzzy regression model, and investigate the performance of the fuzzy regression models, with respect to a certain error measure. Simulation studies and examples show that this model produces less error than the fuzzy regression models studied by many authors, which use the least squares method when the data contains fuzzy outliers. Also this article attempts to optimize the unsteady of the automobile break drum (as the second response variable) getting help from fuzzy regression using least absolute deviation estimators (FLAD) . In the following parts, the amount of optimum effective factor has been calculated via the nonlinear programming model, using one of the multi-objective methods ( LP - Metric).