Feng Sun

Feng Sun

Auto-generated profile, based on publicly available information

Feng Sun

School of Computer and Data Science, Minjiang University, Fuzhou, Fujian, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 December 2019 in E3S Web of Conferences
Verified
Conference Paper
Open access
PDF file
ZZZhao Zhiqin
Zhao Zhiqin
QMQingna Ma
Qingna Ma
XQXu Qian
Xu Qian
SFSun Feng
Sun Feng
Cement concrete pavement Road always damaged under the pressure of traffic, so fast repair technology of cement concrete pavement was an urgent need to be studied. The author introduced the new rapid repair technology of cement concrete pavement from the concrete materials, mixture design, and construction equipment.


17 July 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
FSFeng Sun
Feng Sun
GLGeng Lin
Geng Lin
ZDZhengshan Dong
Zhengshan Dong
JHJianyi Huang
Jianyi Huang
The Minimum Double Roman Domination (MDRD) problem assigns labels $0,1,2$, and$3$ to graph vertices so as to minimize their total weight subject to localprotection requirements. Although these requirements are local, relabeling onevertex may change the feasibility of its entire closed neighborhood; efficientmove evaluation is therefore central to search on general graphs. LS-DRDPaddresses this issue through a penalty-guided multi-start local search for theunweighted MDRD problem. A violated-vertex penalty permits excursions throughinfeasible assignments, while a separate incumbent retains the best feasiblesolution. Two maintained neighborhood counters evaluate one-vertex movesincrementally. Search diversification is supplied by a randomized degree-basedconstructor, and an accumulated-deterioration threshold terminates each searchphase. On Harwell-Boeing graphs, repeated runs are combined with best-valuecomparisons against published genetic algorithm (GA) and ant colonyoptimization (ACO) results. LS-DRDP obtains 90 wins, 20 ties, and 2 lossesagainst GA, versus 47 wins, 31 ties, and 34 losses against ACO; one-sidedWilcoxon signed-rank tests give $p<10^{-15}$ and $p\approx0.0029$,respectively. All 20 runs on each of the 112 instances produce feasiblesolutions. These findings support incremental neighborhood accounting as apractical basis for MDRD local search.