This paper attempts to conduct a comparative life cycle environmental analysis of alternative versions of a product that was manufactured with the use of additive technologies. The aim of the paper was to compare the environmental assessment of an additive-manufactured product using two approaches: a traditional one, based on the use of SimaPro software, and the authors’ own concept of a newly developed artificial intelligence (AI) based approach. The structure of the product was identical and the research experiments consisted in changing the materials used in additive manufacturing (from polylactic acid (PLA) to acrylonitrile butadiene styrene (ABS)). The effects of these changes on the environmental factors were observed and a direct comparison of the effects in the different factors was made. SimaPro software with implemented databases was used for the analysis. Missing information on the environmental impact of additive manufacturing of PLA and ABS parts was taken from the literature for the purpose of the study. The novelty of the work lies in the results of a developing concurrent approach based on AI. The results showed that the artificial intelligence approach can be an effective way to analyze life cycle assessment (LCA) even in such complex cases as a 3D printed medical exoskeleton. This approach, which is becoming increasingly useful as the complexity of manufactured products increases, will be developed in future studies.
Background: Enterprises’ decision-making could be facilitated by properly creating or choosing and implementing demand forecasting systems. Currently, there are more and more advanced forecasting algorithms based on sophisticated technologies such as artificial neural networks and machine learning. The following research paper focuses on a case study of an automotive manufacturer. The main research aim is to propose the proper demand forecasting tool and show the prospects for implementing the mentioned solution. Methods: The research paper contains the statistical analysis of a chosen time series referring to the demanded quantity of the manufactured products. To create forecasts, models based on the following forecasting algorithms were created: ARIMA, ELM (Extreme Learning Machine), and NNAR (Neural Network Autoregressive). All algorithms are based on the R programming language. All algorithms are run in the same time series where the training and testing periods were established. Results: According to the forecasts ex-post errors and FVA (forecasts value-added) analysis, the best fitting algorithm is the algorithm based on ELM. It yields the most accurate predictions. All other models fail to add value to the forecast. Specifically, the ARIMA models damage the forecast dramatically. Such significant magnitudes of negative FVA values indicate that choosing not to forecast and plan based on the sales of the same period of the previous year is a better choice. However, in the case of the ELM model, the forecasts can be worth the time, finance, and human resources put into preparing them. Conclusion: The increased accuracy of ELM forecasts can contribute to optimizing the process of reaching consensus forecasts. While unconstrained statistical forecasts tend to be overridden, not only to produce constrained forecasts incorporating various variables such as calendar events, promotional activities, supply capacity, and operational abilities, they are also overridden by planners to reflect their foreseeing of demand. The proposed solution could also be easily implemented in the resource planning process to improve it. The proposition of the resource planning process supported by the proposed forecasting system is also shown in the following paper using a BPMN 2.0 (Business Process Modelling Notation 2.0) map.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Zbadano możliwości zastosowania sieci neuronowych do modelowania stężenia ozonu przy powierzchni ziemi. Analizę przeprowadzono na podstawie zbioru danych pomiarowych, dla którego we wcześniejszych pracach uzyskano metodami statystycznymi liniowy model regresji wielowymiarowej, uzależniający stężenie ozonu od temperatury, prędkości wiatru i stężenia tlenków azotu NOx. W obliczeniach wykorzystano dane zarejestrowane w 1994 r. na stacji pomiarowej monitoringu powietrza w Krakowie przy ulicy Halickiej. Analizowany zbiór danych stanowiły zmierzone w godzinach nocnych wartości stężeń O3, NOx, temperatury i prędkości wiatru uśrednione w okresach 60-minutowych. Jakość kilku modeli otrzymanych z użyciem sieci neuronowych porównano ze sobą oraz z modelem liniowej regresji wielowymiarowej. Przyjęto dwa kryteria oceny modeli: 1) wartość błędu średniokwadratowego wynikającego z rozbieżności między wyjściem modelu a wartościami rzeczywistymi stężenia O3; 2) wartość współczynnika korelacji dla prostej regresji pomiędzy wyjściem modelu (przewidywaniami) a wartościami rzeczywistymi stężenia O3. Stwierdzono, że modele neuronowe są zdecydowanie lepsze od liniowego. Zweryfikowano pozytywnie zdolność modeli neuronowych do przewidywania wyników dla danych niewykorzystywanych w procesie uczenia, czyli możliwość uogólniania zdobytej wiedzy.
EN
Possibility of neural networks application to the ozone concentration modelling near the ground was examined. The studies were performed based on the measuring data set that was earlier used for generation of multiple linear regression model, conditioning ozone concentration by temperature, wind speed and NOx, concentration. The data gathered in 1994 at the air monitoring station in Cracow were used in the calculations. The analysed data set was built of 60-minutes' averages of temperature, wind speed as well as O3 and NOX concentrations, measured at night. Qualities of some models obtained with neural networks were co m pa red with the multiple linear regression model. Two criteria of model estimation were assumed: 1) the value of mean square error resulting from divergences between model input and real O3 concentration values; 2) the value of correlation coefficient for the best linear fit of the real O3 concentration values to the model input. It was stated that the neural networks models are decidedly better than the linear regression model. Ability of neural models to predict results for the data not used in training process was affirmatively verified.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.