Buffer Monitoring of Critical Chain Projects Based on Support Vector Machine Prediction

PurposeThe critical chain project buffer monitor process addresses uncertainty and variability in project duration. However, classical buffer monitor methods only consider buffer consumption, while the dynamic allocation of buffer zones and the buffer consumption trend of activities are ignored. This paper presents the innovative framework for dynamic monitoring of project buffer which covers the dynamic buffer allocation, predictive analytics of buffer utilization and a new monitoring technique based on control chart graph.Design/methodology/approachFirst, a dynamically buffer allocation model is framed, and buffer zones are given to the activities considering risks. Then, a predictive model amalgamating Bayesian Optimization, Convolutional Neural Networks, and Long Short-Term Memory networks (BO-CNN-LSTM) is framed. Finally, a new buffer monitor framework is constructed that takes into account historical information about buffer usage and utilizes two thresholds derived from control chart theory.FindingsThis approach is empirically tested on a representative agricultural website project in China. The results show that, first, the dynamic buffer allocation makes better use of the project buffer, reduces buffer waste and increases the possibility of timely completion of the project. Second, the BO-CNN-LSTM model predicts better than Long Short-Term Memory (LSTM) and Grey Neural Network Model (GNNM), providing project managers with new management insights and perspectives. Third, the novel monitoring procedure makes the leveraging of historical data possible in the control of the schedule deviations, allowing for more timely interventions in the course of the implementation of the project.Originality/valueA new project buffer monitoring method suitable for uncertain project environments is proposed.

show abstract

Dynamic Prediction Model of Yield of Molten Iron Based on Multi-Head Attention Mechanism

Duan,

Liu,

et al. 2024

ISIJ Int.

View full text Add to dashboard Cite

：Efficient matching of the preheat-numbers of the molten iron ladle with yield of molten iron in a single iron time is conducive to reducing energy waste, so it is of great economic significance to iron and steel enterprises to accurately predict yield of molten iron according to the characteristics of the per furnace. In this article, considering the variation of yield of molten iron and corresponding synergistic parameters depending on furnace conditions, we use the multi-head attention mechanism to capture the Attention-score of the feature parameters, and the weight matrix of the predictor is dynamically adjusted, so that it can assign more training weights to high-value feature parameters in real time according to the change of furnace conditions, efficiently complete the prediction of yield of molten iron. First, twenty characteristics are selected by mutual information method. Then, the influence degree of selected features on yield of molten iron is calculated in real time. Finally, stacked denoising auto encoder (SDAE) is used to train the deep neural network, and we constructed the predictor after adjusting the parameters to optimal. The test results show that the prediction accuracy of proposed model is 95% under the error range of ±50t, which is higher than traditional SVM, DAE, and SDAE model without the multi-head attention mechanism. Ultimately, we analyze and quantify the influence process of 20 factors on yield of molten iron, and develop a dynamic prediction system of yield of molten iron based on proposed model, which has good applicability for the prediction of yield of molten iron and effectively guides the transfer scheduling of molten iron ladles.Keywords：Dynamic prediction model of yield of molten iron; Multi-head attention mechanism; Stacked denoising auto encoder; Transfer scheduling of molten iron ladles;

show abstract

Buffer Monitoring of Critical Chain Projects Based on Support Vector Machine Prediction

Cited by 4 publications

References 35 publications

Monitoring and controlling engineering projects with blockchain-based critical chain project management

Monitoring and controlling engineering projects with blockchain-based critical chain project management

A hybrid approach based on critical chain for project buffer consumption prediction and monitoring

Dynamic Prediction Model of Yield of Molten Iron Based on Multi-Head Attention Mechanism

Contact Info

Product

Resources

About