We propose a new multi-layer explosion strategy inspired by various explosion patterns of real reworks to accelerate reworks algorithm (FWA). Each rework individual conducts multiple explosions to explore a local fitness landscape carefully instead of a single layer explosion used in canonical FWA. In the proposal, each firework individual generates a small number of sparks in the first layer randomly, and then the generated sparks conduct the second layer explosions to generate new diverse sparks. These new sparks repeat the above operations until the number of this iteration reaches the predefined maximum layer number. Theoretically, the number of explosion layers can be set to any positive integer, and the proposed strategy expects to generate various potential sparks using the multi-layer explosion strategy without changing the total number of generated sparks. The proposed strategy can combine with not only basic FWA but also other versions of FWA algorithms easily and replace their corresponding explosion operations to develop a new version, multi-layer explosion-based FWA.

Fireworks Algorithm

In real world, a firework is launched into the sky then many sparks are generated around it. A firework explosion can be abstracted as a local search model around a specific point. FWA simulates the explosion process iteratively to find the optimal candidate solution.

Conventional FWA has two types of fireworks, better rework and poor firework. A better firework may close to an optimum so that many sparks are generated around it within a smaller range, while a poor firework may far from the optimum so less sparks are generated, and search range should be larger. Algorithm 1 shows the general own of FWA mainly consisting of three operations: explosion, mutation and selection

Algorithm 1

• Be general framework of FWA.
• Initialize n fireworks randomly.
• Evaluate the fitness of each firework.
• While termination condition is not satisfied do
• Generate sparks for each firework.
• Generate Gaussian mutation sparks.
• If sparks are generated outside search area, then
• Use a mapping rule to bring it back to search space.
• End if
• Evaluate the fitness of generated sparks.
• Select n new fireworks for next generation.
• End while
• Output found feasible solution.

Prospective academicians and scientists are encouraged to utilize this opportunity to get their articles reviewed, processed and published at relatively faster pace and at lower charges. Journal accepts original manuscripts in the form of research articles, review articles, Clinical reviews, commentaries, case reports, perspectives and short communications encompassing all aspects of Reproductive Sciences for publication in open access platform.

All the manuscript published by International Journal of Swarm Intelligence and Evolutionary Computation is available freely online immediately after publication without any subscription charges or registration. Manuscripts can be uploaded at:  www.longdom.org/submissions/international-swarm-intelligence-evolutionary-computation.html

After submission, an acknowledgement with manuscript number is sent to the corresponding author within 7 working days. A 21-day window time frame is allotted for peer-review process wherein multiple experts are contacted.

Author proof is generated within 7 working days after the acceptance decision.

Benefits on Publication: Permanent free access to your article upon publication ensures extensive global reach and readership. A strict single blinded peer review process is being followed to assess each submission and to enrich the quality. All published articles are indexed in the leading indexing services.

Easy Article Sharing: Our open access enables you to share your article directly with colleagues through email and on social media via a single link, permitting third party reuse with appropriate citation in addition to the retention of content copyright by the author.

Global Marketing: Through promotion in a targeted global email announcement or press release, your article will be seen by thousands of the top-most thought-leaders in your field.

Social Media Exposure: Extended reach for your article through links on Twitter accounts provides maximum visibility worldwide.

Reprints: Distribute your work to colleagues and at conferences as we provide hard copy color reprints of your article on order.

Media Contact:
Sarah Rose
Journal Manager
International journal of swarm intelligence and evolutionary computation
Email: evolcomput@journalres.org