Huang De-cai

Uncertainty network planning technique using the connection number a+bi of set pair analysis

The paper uses the connection number a+bi of set pair analysis theory (SPA) to express and process the uncertainties of the time limit caused by fuzzy, random and indeterminate factors in network planning of a large-scale project, and researches the managing and controlling methods of the network planning. The results in the paper illustrate that the network planning and controlling method using the connection number a+bi can provide more useful information such as primary critical paths, secondary critical paths and tertiary critical paths etc. The new uncertainty network planning technique is more flexible and more suitable to the practical application of project than the traditional one.

Research of subjective trust comprehensive evaluation model based on multi-element connection number

Trust management model is fundamental for information security in the open network environment. Subjective trust has fuzziness and uncertainty dual characteristics. The multi-element connection number is introduced into subjective trust management, and then a new representation method for subjective trust based on multielement connection number was proposed. This method takes into account uncertainty and fuzziness of subjective trust, overcomes a shortcoming that is expressing strictly the fuzziness concept using a precise, unique membership function in fuzzy math, and provides a new valuable way for studying subjective trust evaluation.

A heuristic for optimal job scheduling problem with a common due window on parallel and identical machines

The paper is concerned with the job scheduling problem with a common due window. After giving an optimal algorithm for the single machine problem, a mathematical model for the similar problem on parallel and identical machines is presented. Because the job scheduling problem on parallel and identical machines, which may be NP-complete, is much more complex than that on a single machine, a heuristic algorithm is given to find an approximate solution after two important theorems are proved. The numerical example illustrates that the heuristic algorithm is very effective in obtaining a near-optimal solution.

A heuristic algorithm for minimizing the range of lateness and makespan on non-identical multiprocessors

A generalized multi-objective model is presented for minimizing the range of lateness and makespan on a multiprocessors job scheduling problem. It is proved that this job scheduling problem is an NP complete problem. So a heuristic algorithm is given to solve the scheduling problem. The numerical example illustrates that the heuristic algorithm is designed very well and can find very good approximate optimal solution.

Modified DMC method and its realization based on BP neural networks

Presents a modified dynamic matrix control (DMC) algorithm and its realization based on BP neural networks with higher nonlinear mapping and learning ability, predictive precision is increased and the control quality is improved by this control algorithm. In addition, the predictive control algorithm is extended to nonlinear systems. The results of simulation show that this algorithm is simple and applicable; its robustness is high; the transient and steady-state quality and the oscillation of the control variable are improved.