Optimization of steel demand forecasting with complex and uncertain economic inputs by an integrated neural network–fuzzy mathematical programming approach

Azadeh, A.; Neshat, Najmeh; Mardan, Ehsan; Saberi, Morteza

doi:10.1007/s00170-012-4221-1

Cited by 7 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the potential drivers for secondary copper supply are captured by indices of industrial activity, prices, and monetary conditions. They were chosen based on a survey of econometric literature pertinent to metals markets and aim to be comprehensive in their coverage of activities that may influence the supply and demand of copper more generally (Azadeh et al, 2013;Elshkaki et al, 2005;Finnveden, Ekvall, 2013, Reck andGraedel, 2012). The choice of independent variables was also based on our interest in reflecting scrap generation rather than scrap consumption.…”

Section: Methodsmentioning

confidence: 99%

Econometric modeling of recycled copper supply

Ueland

Olivetti

2017

Resources, Conservation and Recycling

View full text Add to dashboard Cite

The supply of recycled material depends on historic consumption, i.e. what constitutes scrap available today originates from previously made products. Analytical tools, such as materials flow analysis, use this observation to estimate scrap metal flows. The supply of recycled metal also depends on changing economic conditions, e.g. metal consumption rates correlate with changes in gross domestic product. We use an autoregressive distributed lag approach to model the supply of recycled copper as a complementary approach to material flow analysis. We find that both industrial activity and world GDP correlate with total scrap supply, with limited dependence on copper price. We also develop independent models for direct remelt (higher quality) and refined (lower quality) scrap. A 1% increase in industrial production leads to a 2.1% increase in higher quality scrap quantity, while a similar increase in world GDP leads to a 1.4% increase in lower quality scrap. Based on this model dependence, we suggest that a recycling policy aimed at increasing recycling through the use of subsidies, taxes or price incentives should be directed towards the low-end segment of the scrap market and there it may still only have limited impact.

show abstract

Section: Methodsmentioning

confidence: 99%

Econometric modeling of recycled copper supply

Ueland

Olivetti

2017

Resources, Conservation and Recycling

View full text Add to dashboard Cite

show abstract

“…Traditionally, AI-based algorithms such as artificial neural network [14][15][16][17], particle swarm optimization algorithm [18,19], wavelet neural network [20][21][22], and some coupling algorithms [23][24][25] were used to solve the problems with nonlinearity and fuzziness. In fact, as a perfect integration approach of wavelet transform and neural network, wavelet neural network could solve complex nonlinear systems in a manner of high nonlinear mapping ability, fine local characteristic, and better generalization performance compared with traditional neural network [26][27][28].…”

Section: Wavelet Neural Networkmentioning

confidence: 99%

A Novel Approach for Shearer Cutting Load Identification through Integration of Improved Particle Swarm Optimization and Wavelet Neural Network

Wang

Tan

et al. 2014

Advances in Mechanical Engineering

View full text Add to dashboard Cite

In order to accurately identify the change of shearer cutting load, a novel approach was proposed through integration of improved particle swarm optimization and wavelet neural network. An improved updating strategy for inertia weight was presented to avoid falling into the local optimum. Moreover, immune mechanism was applied in the proposed approach to enhance the population diversity and improve the quality of solution, and the flowchart of the proposed approach was designed. Furthermore, a simulation example was carried out and comparison results indicated that the proposed approach was feasible, efficient, and outperforming others. Finally, an industrial application example of coal mining face was demonstrated to specify the effect of the proposed system.

show abstract

“…x@ \w@Qt C=aq]= R= xO=iDU= =@ Oqwi hQYt u= R}t QO C= Q}}eD OvwQ |UQQ@ COW x} Q_v 2015 [20] [21] w} S O}rwD Q}eDt uDiQo Q_v QO =@ u= Q}= QO s=N Oqwi hQYt |v}@V}B |@Ya |=yxm@W 2013 [22] [25] u= Q=mty w nvw Vwy |=yQ= R@= |Q}oQ=m x@ w i;j (w i;j ) = 8 > > > > < > > > > : 1 b i;j a i;j (w i;j a i;j ) if a i;j w i;j b i;j ; 1 b i;j c ij (wi;j ci;j) if bi;j wi;j ci;j; 0 otherwise; (12) wi;j |=yQDt= Q=B =@ xm CU}= |R=i xawtHt C} w[a`@=D w(wi;j) xm|Qw]x@ w |FrFt |R=i O=Oa= CQwYx@ |R=i |=yQDt= Q=B R= xO=iDU= =@ p=L "OwW|t XNWt j@=]tx t;j = g(w 0;j + p P i=1w i;j :y t i ) C} w[a`@=D 'VQDUo pY= u}vJty [31] "Ot; Oy=wN CUO@ p}P X t;j (x t;j ) = 8 > > > > > > > > > < > > > > > > > > > :…”

mentioning

confidence: 99%