Zhiqiang Xie

A new algorithm for complex product flexible scheduling with constraint between jobs

In order to simplify the complex product flexible scheduling problem with constraint between jobs, a new hierarchical scheduling algorithm based on improved processing operation tree is presented. Aiming at the routing problem, short-time strategy and machine-balance strategy are adopted to assign each operation to a machine out of a set of machines. And in order to solve the sequencing problem, the allied critical path method is first adopted to confirm the scheduling sequence of operations, and then operations are divided into dependent operations and independent ones according to their characteristics. For the dependent operations, forward greedy rule is adopted in order to make the completion time of operation as soon as possible and the scheduling algorithm of shortening idle time is adopted by analyzing the characteristics of the independent operations. Experiment shows that the proposed algorithm solves for the first time the complex product flexible scheduling problem with constraint between jobs.

Notice of Retraction Study on integrated algorithm of complex multi-product flexible scheduling

In order to optimize the processing of complex multi-product flexible scheduling, a simplified method which converts complex multi-product into a virtual single-product is adopted. The complex multi-product flexible scheduling problem is transformed into the virtual single-product flexible scheduling problem. Aiming at the virtual single-product flexible scheduling problem, an integrated algorithm is proposed. This algorithm firstly adopts the allied critical method to confirm the scheduling sequence of operations. Then each operation is simulation processed by the sequencing algorithm on a set of capable machines. Calculate the virtual completion time of this operation and choose the machine which has the minimum virtual completion time. The routing sub-problem and sequencing sub-problem are solved at the same time by the proposed algorithm. Experiment shows that the proposed algorithm can solve complex multi-product flexible scheduling problem with constraint between jobs, it also can solve simple multi-product flexible job-shop scheduling problem with non-restraint between jobs and obtain satisfying result.

Notice of Retraction Study on Confirming Increasable Bottleneck Device in Complex Product Dynamic Scheduling

Aiming at the problem that the machine is single and some machine has effect on the processing efficiency in complex product dynamic scheduling, based on the conceptions of bottleneck device classification and increasable bottleneck device, an algorithm which can confirm the increasable bottleneck device is put forward by calculating the total time of parallel operations and by analyzing the number of immediate predecessors. Calculating the total time of parallel operations for a device can confirm the shrinkable times of increasing this device. When devices have the same total time of parallel operations, choosing the device which has the most number of immediate predecessors can make more operations be processed as soon as possible. The verification and comparison of example show that the aim of improving the production efficiency is achieved by adding the increasable bottleneck device. The algorithm is simple and it can confirm the increasable bottleneck device quickly in complex product dynamic scheduling.

Study on Job Shop Scheduling with Many Function-Same Machines

Aiming at some operations of job can be assembled by any machine selected from the function-same machine set, a secondary optimization algorithm for operations on function-same machines is put forward: the machine which can make the completion time of operation early will be selected according to a dispatching rule (earliest due date, EDD), and operations are arranged in the right position of machine selected, then each operation can be assembled as soon as possible, and the total processing time of job can be shortened as soon as possible. To independent operations optimum scheduling algorithm is adopted: the inserting position of operations will be confirmed according to the margin that processing time of operations is subtracted from idle time of machine, and the operation will be inserted in the position with minimal margin. The experiment shows that the new algorithm without increasing complexity has better scheduling result for job shop scheduling problem with many function-same machines.

An Algorithm for Complex Product Dynamic Integrated Flexible Scheduling

Aiming at the dynamic scheduling problem with many products, a virtual processing operation tree is constructed by products with different arriving times. The multi-product dynamic flexible scheduling problem is converted into the virtual single-product flexible scheduling problem. A new hierarchical scheduling algorithm for solving dynamic flexible scheduling problem with constraint between jobs is present in this study. Aiming at the routing problem, short-time strategy and machine-balance strategy are adopted to assign each operation to a machine out of a set of capable machines. And in order to solve the sequencing problem, operations are divided into the dependent operations and the independent ones. To the dependent operations, forward greedy rule is adopted in order to make the completion time of operation as early as possible and the scheduling algorithm of shorten idle time is adopted by analyzing the characteristics of the independent operations. Experiment shows that the proposed algorithm can solve firstly the complex product dynamic flexible scheduling problem with constraints between jobs.

Notice of Retraction The algorithm of confirming increasable bottleneck device in dynamic integrated scheduling based on static parallel time

Aiming at the problems that there is a big difference between the parallel total time of operations and real parallel time in processing and assembly integrated scheduling of dynamic complex product to confirm increasable bottleneck device, and an algorithm which can confirm the increasable bottleneck device in dynamic integrated scheduling based on static parallel time is presented. Firstly, a processing tree of virtual product, consisting of remaining (remaining operations is included) and new products, is constructed. Then according to the processing tree of virtual product, the algorithm calculates operations starting and ending times, according to overlapping projection, calculates the total static parallel time, and selects a device whose total static parallel time is the largest as an dynamic increasable bottleneck device; when new products arrive to be processed, the remaining operations have priority to schedule, the rest of operations adopt ACPM to confirm scheduling order. The analysis and example show that the algorithm confirming the dynamic increasable bottleneck device is justifiable.

A Dynamic Scheduling Algorithm for Complex Product with Batching Machines

At present, scheduling research with batching machines mainly solves scheduling problem without constraint among operations. There is no effective scheduling method for complex product with batching machines. To that end, a new algorithm is proposed to solve dynamic scheduling problem for complex product with batching machines. And the maximum lot-size of the batching machines is two. First, based on the feature of the structure of the processing tree which is provided with complex product, the algorithm adopts Priority strategy, Short-time strategy and Long-path strategy to schedule operations. Second, when the operations belong to batching machine, the algorithm adopts Long-path strategy, Critical-time waiting strategy and Short-time of precursor strategy to schedule operations. Theoretical analysis and examples show that the algorithm can solve dynamic scheduling problem for complex product with batching machines and the maximum lot-size of the batching machines is two. In addition, the complexity is not more than quadratic polynomial.

Notice of Retraction Complex multi-products dynamic scheduling algorithm with sequence-dependent setup times

The problem of considering the setup times in scheduling with non-sequence constraint between operations has been researched, but there are no researches on the problem of considering the setup times in processing and assembly integrated scheduling of complex multi-product dynamic scheduling with sequence constraint between operations. So, a complex multi-product dynamic scheduling algorithm is proposed, considering sequence-dependent setup times. The algorithm constructs the virtual manufacturing tree in the way of root alignment and converts the complex multiproduct dynamic scheduling problem to the complex single product scheduling problem. After the scheduling sequence of operations is determined, the starting time of each operation is determined by dynamically inserting setup time once. As the 1 proposed algorithm avoids several movements of the scheduled operation when the setup time is inserted, the time complexity is sharply reduced. Theoretical analysis and examples show that the proposed algorithm not only can solve the complex multi-product dynamic scheduling problem with sequence-dependent setup times, but also is simple and practical.

Method for Confirming Increasable Bottleneck Device Based on Study of Parallel Operation

Common job-shop scheduling problem is based on that every processing device is single so there is usually certain kind of device which becomes the scheduling bottleneck and affects the processing efficiency of production most. Adding the bottleneck device is a simple method for solving bottleneck problem, but the key problem includes two sides. The first is to find the bottleneck device and the second is to confirm that it is such an increasable device that the processing efficiency can have an impressive promotion after a same device is added. The method for confirming the increasable bottleneck device is by judging whether the device has the total longest processing time of parallel operations. An accurate and fast method for judging parallel operations was put forward by studying parallel operations. Analysis and examples validate the method proposed can confirm the increasable bottleneck device quickly.