Papers by Author: B. Kareem

Abstract: Attainment of accuracy in raw materials mix for production processes has been the major problem in many production organizations in developing countries. Effects of unequal proportions of material required for a production process are not fully considered in the past studies. This study modeled the raw material requirements in a production process using proportionality based mixed linear programming approach. The objective was to find optimal mix of raw materials for the production of a unit tonnage of a product. The performance of the model was tested by comparing it with alternatively formulated model based on conventional material mix. This model was tested using a cement production system from which blasted limestone, crushed limestone; raw meal, gypsum, red alluvium, clinker and coal were used as raw materials. The conventional method of mixing materials led to 50 % surplus and shortage of materials in the process as compared with the new scheme. The model would be a good tool for accurate prediction of quantity of the raw material required in the production process.Nomenclature, the proportion of material used per ton of the processed product, the quantity of material in processed product (ton), the material proportional per ton of the product, andthe total quantity of materials needed per ton of the product,..., the counter for material type,..., the counter for material proportionality variant

Abstract: Queuing model has been discussed widely in literature. The structures of queuing systems are broadly divided into three namely; single, multi-channel, and mixed. Equations for solving these queuing problems vary in complexity. The most complex of them is the multi-channel queuing problem. A heuristically simplified equation based on relative comparison, using proportionality principle, of the measured effectiveness from the single and multi-channel models seems promising in solving this complex problem. In this study, six different queuing models were used from which five of them are single-channel systems while the balance is multi-channel. Equations for solving these models were identified based on their properties. Queuing models’ performance parameters were measured using relative proportionality principle from which complexity of multi-channel system was transformed to a simple linear relation of the form = . This showed that the performance obtained from single channel model has a linear relationship with corresponding to multi-channel, and is a factor which varies with the structure of queuing system. The model was tested with practical data collected on the arrival and departure of customers from a cocoa processing factory. The performances obtained based on average number of customers on line , average number of customers in the system , average waiting time in line and average waiting time in the system, under certain conditions showed no significant difference between using heuristics and analytical models.

Abstract: Conventional tool monitoring instruments are usually costly to acquire. The instruments are inadequate for real time wear measurement in the uncertain environmental conditions of developing countries. A modeling approach relating relevant parameters causing wear on cutting tools’ flank will be useful in predicting wear in machining operations. Mild steel turning experiment was carried out on the lathe using selected High Speed Steel (HSS) and High Carbon Steel (HCS) single point cutting tools. Speed, feed, and time of machining were varied accordingly, while running with, and without coolant. Hardness of the cutting tips of the tools was measured using Rockwell, R hardness tester. The corresponding speed, feed and time of machining were also noted. Turning operation was continued until the tool was totally blunt. At this stage recorded values of hardness, time of machining, feed and speed were modeled using multiple regression technique, with and without cutting fluid. The resulting models were strongly in agreement with the measured values. Therefore, the model is a good predictor of flank wear for the selected tools commonly used in developing countries. The findings showed that wear of the cutting tools can be predicted during machining at predetermined cutting conditions.

Abstract: In this study, investigations were carried out on the corrosion behaviour of low carbon steel (mild steel) in the selected media which are cassava extracts, sodium chloride solution (brine), lubricating oil, distilled water and atmospheric air. The test pieces, after chemical analysis were machined to similar dimensions. Each of the specimens was immersed into equal volumes of the selected media except one for the atmospheric air which was placed in the open air. The weight and hardness measurements were carried out every five days for a period of twenty-five days and loss in weight and hardness of the specimens were determined in succession. The results obtained showed that mild steel was more susceptible to corrosion in cassava extracts, sodium chloride and distilled water, than in lubricating oil and atmospheric air. Follow up statistical test results showed that loss in weight is significant for cassava extracts when compared with other media.