Dealing with missing usage data in defect prediction: A case study of a welding supplier

2021 ◽  
Vol 132 ◽  
pp. 103505
Author(s):  
Milot Gashi ◽  
Patrick Ofner ◽  
Helmut Ennsbrunner ◽  
Stefan Thalmann
Keyword(s):  
2019 ◽  
Vol 53 (18) ◽  
pp. 2505-2515 ◽  
Author(s):  
Benoit Landry ◽  
Pascal Hubert

A numerical model was developed to predict the defect formation during processing of compression moulded discontinuous long fibre carbon/polyether ether ketone composites. The model inputs are the material's temperature-dependant properties (through-thickness modulus and thermal shrinkage), the temperature distribution of the part during cooling and the applied moulding pressure. The material properties of carbon/polyether ether ketone prepreg were measured during cooling from melt using thermal analyses. The model was employed to identify regions on manufactured panels where pressure could be lost during cooling, which are prone to defect formation. Validation was performed by comparing the predicted defect areas against those found on flat panels moulded at pressures ranging from 10 to 110 bar. The model was then employed in a case study to show the importance of the cooling strategy in order to prevent defects on complex-shape components.


2017 ◽  
Vol 12 (1) ◽  
pp. 109
Author(s):  
Laura Costello

A Review of: Carrico, S.B., Cataldo, T.T., Botero, C., & Shelton, T. (2015). What cost and usage data reveals about e-book acquisitions: Ramifications for collection development. ALCTS, 59(3). Retrieved from https://journals.ala.org/lrts/article/view/5752/7199 Abstract Objective – To compare e-book cost-usage data across different acquisitions styles and disciplines. Design – Case study. Setting – A public research university serving an annual enrollment of over 49,000 students and employing more than 3,000 faculty members in the Southern United States. Subjects – Cost and usage data from 15,006 e-books acquired by the Library through packages, firm orders, and demand-driven acquisitions. Methods – Data was collected from publishers and vendors across the three acquisitions strategies. Usage, cost, and call number information was collected for the materials purchased via firm order or demand driven acquisitions and these were sorted into disciplines based on the call number assigned. Discipline, cost, and use were determined for each package collection as a whole because information on individual titles was not provided by the publishers. The authors then compared usage and cost across disciplines and acquisitions strategies. Main Results – Overall, e-books purchased in packages had a 50% use rate and an average cost per use of $3.39, e-books purchased through firm orders had a 52% use rate and an average cost per use of $22.21, and e-books purchased through demand driven acquisitions had an average cost per use of $8.88 and 13.9 average uses per title. Package purchasing was cost effective for science, technology, engineering, and mathematics (STEM) materials and medicine (MED) materials. Demand driven acquisition was a particularly good strategy for humanities and social sciences (HSS) titles. Conclusion – There are differences between the acquisitions strategies and disciplines in cost and use. Firm orders had a higher cost per use than the other acquisitions strategies.


Sign in / Sign up

Export Citation Format

Share Document