Search

Preventive maintenance (PM) can benefit systems having an increasing failure rate. However, PM is often scheduled in order to minimize the maintenance cost or to comply with production planning requirements or instruction from the equipment suppliers. Given that maintenance actions affect process time variability and resource utilization, such maintenance planning criteria may cause other adverse effects on the operational performance of the manufacturing system such as an increase in Work In Process (WIP) and cycle time. In this paper some approximate queueing models are utilize to asses the impact of PM interval on WIP. It is shown that WIP value is strongly influenced by the PM interval and that maintenance intervals corresponding to a minimum maintenance cost or minimum WIP can be quite different. This kind of analysis can help in making more informed decisions involving WIP and cost trade-offs. [ABSTRACT FROM AUTHOR]

This paper studies the simultaneous scheduling of landside container handling operations in a container terminal. Issues addressed include scheduling the sequence of loading (unloading) of containers to (from) the vessels from (to) the quayside, assigning trucks to transport containers between quayside and yard side, and scheduling operations of yard cranes in different yard zones. A mathematical model describing the characteristics of the problem is developed. The objective is to minimize the total completion time for handling all the containers under consideration in the terminal. As optimizing the landside container handling operations in a terminal is known to be NP-hard, a new genetic algorithm with the selection process based on the principle of simulated annealing is developed in this paper to solve the problem. Comparison of the respective results obtained by using the proposed genetic algorithm, the canonical genetic algorithm and the simulated annealing algorithm clearly shows that the total completion times obtained by the proposed algorithm are 12%-18% shorter than that obtained by GA and SA, and the computing times of GA-SA are only 50% of that of GA. The proposed genetic algorithm is indeed superior to the other two algorithms. [ABSTRACT FROM AUTHOR]

Sequence-dependent setup times can have a major influence on manufacturing systems performance. Although much investigation has been given to this issue, the role of setup savings from the perspective of the workload control concept has hardly received attention. This paper reports a simulation study on this matter. The study evaluates the potential for savings in setups, dependent on the level of workload in the shop, for two alternative strategies, namely considering setup times centrally, within the release decision or locally, within the dispatching decision. These strategies are compared and assessed on the basis of two performance measures: time in system and standard deviation of job lateness. Results show that shop configuration is an important factor to choose the strategy to adopt. Moreover, the level of workload is also critical performance of the strategies. [ABSTRACT FROM AUTHOR]

Effective and efficient scheduling of yard crane operations is essential to guarantee a smooth and fast container flow in a container terminal, thus leading to a high terminal throughput. This paper studies the problem of scheduling yard cranes to perform a given set of loading and unloading jobs with different ready times in a yard zone. In particular, the inter-crane interference between adjacent yard cranes which results in the movement of a yard crane being blocked by adjacent yard cranes is studied. The objective is to minimize the sum of yard crane completing times. Since the scheduling problem is NP-complete, a new hybrid optimization algorithm combining the techniques of genetic algorithm and tabu search method (GA-TS) is proposed to solve the challenging problem. Two new operators, namely the Tabu Search Crossover (TSC) and the Tabu Search Mutation (TSM), are introduced into the proposed algorithm to ensure efficient computation. A set of test problems generated randomly based on real life data is used to evaluate the performance of the proposed algorithm. Computational results clearly indicate that GA-TS can successfully locate cost-effective solutions which are on average 20% better than that located by GA. Indeed, the proposed hybrid algorithm is an effective and efficient means for scheduling yard cranes in computer terminals. [ABSTRACT FROM AUTHOR]