Search

In vendor managed inventory (VMI) operations, it is necessary for both the suppliers and the customers to reach an agreement on the maximal and minimal inventory level. In practice, in VMI, inventory levels are decided manually by planning personnel, based on their experience and past operating records. From a system optimization point of view, the determination of the upper and lower inventory levels in storage areas is complicated, necessitating analysis of many factors. Consequently, it may not be possible to obtain the best planning results, to determine and maintain the optimum levels of inventory. In this study, we utilize network flow techniques to build a replenishment model to deal with the upper-and-lower inventory level control problem, with the objective of minimizing the total cost in short-term operations, subject to inventory level control and other operating constraints. The model is formulated as an integer network flow problem with side constraints and is characterized as NP-hard in terms of optimization. To efficiently and effectively solve the large-scale problems that occur in the real world, a solution algorithm is also developed. Finally, numerical tests are conducted using real data from a major retail company in northern Taiwan. The results demonstrate the usefulness of the proposed model and solution algorithm for replenishment planning in actual practice.

Stochastic disturbances occurring in real world operations could have a significant influence on the planned routing and scheduling results for inter-school buses. An optimal schedule might therefore lose its optimality when applied in actual operations where stochastic disturbances occur. In this study we employ network flow techniques to construct an inter-school bus routing and scheduling model, considering the characteristics of inter-school buses for class attendance, under stochastic travel times. The model is formulated as a special multiple commodity network flow problem which is characterized as NP-hard. To efficiently solve realistically large problems a problem decomposition and variable fixing method, coupled with the use of CPLEX mathematical programming solver, is employed to develop a solution algorithm. Finally, computational tests are performed using data from an allied university system in Taiwan. The test results show that the model and the solution algorithm are good and therefore could be useful for actual operations.

In order to optimize the structure of high temperature superconducting coils for linear motor applications, three separate coils with different shapes made of Re-BCO coated conductor were studied: circular shaped single pancake, circular shaped double-pancake and racetrack shaped single-pancake. The thrust and vertical forces of the three coils above a conventional flat linear three-phases winding were investigated experimentally. With the aid of the experimentally obtained values, it was found that single-pancake coil in the shape of a racetrack was the best selection for a flat single-sided linear motor system. Studies were also made on the frequency characteristics of the vertical force of the racetrack shaped single-pancake coil.

High-Tc superconducting (HTS) bulk samples used for a maglev system can be re-magnetized after being subjected to another applied magnetic field which may influence the levitation performance of a HTS bulk over the permanent magnetic guideway (PMG). In order to clarify the relationship between the HTS bulk’s re-magnetization and its levitation performance, a YBCO bulk was re-magnetized at different values and its levitation performance studied experimentally above a permanent guideway (PMG) system. The results show that the increase of trapped magnetic flux inside the HTS bulk is closely related to the magnitude and direction of the applied field during the re-magnetization process. Furthermore, the levitation force changes on the HTS-PMG system affected by the re-magnetized HTS bulk are discussed allowing for optimization improvements to the maglev system load capabilities.

The distribution of and relationship between F-actin and G-actin were investigated in pollen grains and pollen tubes of Lilium davidii Duch. using a confocal laser scanning microscope after fluorescence and immunofluorescence labeling. Circular F-actin bundles were found to be the main form of microfilament cytoskeleton in pollen grains and pollen tubes. Consistent with cytoplasmic streaming in pollen tubes, there were no obvious F-actin bundles in the 10- to 20-microm tip region of long pollen tubes, only a few short F-actin fragments. Labeling with fluorescein isothiocyanate (FITC)-DNase I at first established the presence of a tip-focused gradient of intracellular G-actin concentration at the extreme apex of the tube, the concentration of G-actin being about twice as high in the 10- to 20-microm region of the tip as in other regions of the pollen tube. We also found that the distribution of G-actin was related negatively to that of the F-actin in pollen tubes of L. davidii. Caffeine treatment caused the G-actin tip-focused gradient to disappear, and F-actin to extend into the pollen tube tip. Based on these results, we speculate that the circular F-actin bundles may be the track for bidirectional cytoplasmic streaming in pollen tubes, and that in the pollen tube tip most of the F-actin is depolymerized into G-actin, leading to the absence of F-actin bundles in this region.