Fusion of Activation Functions: An Alternative to Improving Prediction Accuracy in Artificial Neural Networks

The purpose of this study was to address the challenges in predicting and classifying accuracy in modeling Container Dwell Time (CDT) using Artificial Neural Networks (ANN). This objective was driven by the suboptimal outcomes reported in previous studies and sought to apply an innovative approach to improve these results. To achieve this, the study applied the Fusion of Activation Functions (FAFs) to a substantial dataset. This dataset included 307,594 container records from the Port of Tema from 2014 to 2022, encompassing both import and transit containers. The RandomizedSearchCV algorithm from Python’s Scikit-learn library was utilized in the methodological approach to yield the optimal activation function for prediction accuracy. The results indicated that “ajaLT”, a fusion of the Logistic and Hyperbolic Tangent Activation Functions, provided the best prediction accuracy, reaching a high of 82%. Despite these encouraging findings, it’s crucial to recognize the study’s limitations. While Fusion of Activation Functions is a promising method, further evaluation is necessary across different container types and port operations to ascertain the broader applicability and generalizability of these findings. The original value of this study lies in its innovative application of FAFs to CDT. Unlike previous studies, this research evaluates the method based on prediction accuracy rather than training time. It opens new avenues for machine learning engineers and researchers in applying FAFs to enhance prediction accuracy in CDT modeling, contributing to a previously underexplored area.

Theoretical study of fusion reactions ^(32)S + ^(94,96)Zr and ^(40)Ca + ^(94,96)Zr and quadrupole deformation of ^(94)Zr

The dynamic coupling effects on fusion cross sections for reactions^(32)S + ^(94,96)Zr and ^(40)Ca + ^(94,96)Zr are studied with the universal fusion function formalism and an empirical coupled channel(ECC) model. An examination of the reduced fusion functions shows that the total effect of couplings to inelastic excitations and neutron transfer channels on fusion in ^(32)S +^(94)Zr(^(40)Ca +^(94)Zr) is almost the same as that in ^(32)S +^(96)Zr(^(40)Ca +^(96)Zr). The enhancements of the fusion cross section at sub-barrier energies due to inelastic channel coupling and neutron transfer channel coupling are evaluated separately by using the ECC model. The results show that effect of couplings to inelastic excitations channels in the reactions with94 Zr as target should be similar as that in the reactions with ^(96) Zr as target. This implies that the quadrupole deformation parameters β_2of ^(94)Zr and^(96) Zr should be similar to each other.However, β_2 's predicted from the finite-range droplet model, which are used in the ECC model, are quite different. Experiments on^(48) Ca +^(94)Zr or^(36) S +^(94)Zr are suggested to solve the puzzling issue concerning β_2for^(94)Zr.