Finding the k shortest paths by ripple-spreading algorithms

Co-Evolutionary Location-Routing Model of Medical Isolation Areas for Major Public Health Emergencies Considering the Uncertainty of Future Traffic Environment

“…Different from the traditional route-planning method, IRSA is a time-oriented algorithm. Inspired by the natural phenomenon of the spreading of ripples on the water surface ( 36 , 37 ), IRSA develops a special ripple relay race to find the best solution. In the ripple relay race, when the ripples start to spread out from the origin to the adjacent nodes, new ripples will be triggered and continue to spread out once a ripple reaches the node, and the ripple relay race will be terminated when all destination nodes are triggered.…”

Intelligent Tourism Recommendation Algorithm based on Text Mining and MP Nerve Cell Model of Multivariate Transportation Modes

“…Different tourist sights have different function attributes, which will bring discrepancy on tourists' satisfaction. Then, the time schedule, travel cost and the choice of transportation mode will also influence the tourist sight and tour route recommendation [21]. The classification and quantity of the recommended tourist sight are not only determined by tourists' interests, but also by tourist sight distribution, time schedule and travel cost.…”

Finding all Pareto optimal paths by simulating ripple relay race in multi-objective networks

“…Inspirations from the nature have successful helped researchers to develop many effective multi-agent-based evolutionary computation methods [1] , [2] . Ripple-spreading models and algorithms are relatively new in the family of nature-inspired multi-agent-based methods, but they have shown potentials in studying some problems (e.g., sequencing aircraft [3] , modeling complex networks [4] , simulating epidemic dynamics [5] , finding the first shortest path in static [6] and dynamical environments [7] , calculating the k shortest paths without [8] and with time-windows [9] ), and they suggest a possibility of new stream of optimization methods, distinguished from both classical deterministic method stream (e.g., A * algorithm [10] , Dijkstra's algorithm [11] and label correcting algorithm [12] ) and stochastic method stream (e.g., genetic algorithm [13] , particle swarm optimization [14] , ant colony optimization [15] and firefly optimization [16] ). Usually, deterministic methods have optimality guarantee but are not flexible in modification, while stochastic methods often output non-optimal solutions but are easy to modify for various applications.…”

“…As shown in [6] and [7] , RSA, as a newly reported, nature-inspired, multi-agent-based but deterministic algorithm, exhibits some advantages against traditional Dijkstra's algorithm and A * algorithm, particularly when it comes to resolve some complicated path optimization problems. For example, when route networks with time windows or dynamical routing environments are concerned, RSA has demonstrated a good modification flexibility and computational efficiency [7] , [8] . This paper, with finding complete Pareto front as the focus, is first interested in exploring the possibility of applying RSA to target MOPOP with static route networks, and then, it goes on to extend to MOPOP with dynamical route networks, which are more complicated and have rarely been touched in those MOPOP literatures such as [45] - [49] .…”