A Data-Driven Framework to Predict Fused Filament Fabrication Part Properties Using Surrogate Models and Multi-Objective Optimisation

In additive manufacturing (AM), due to large number of process parameters and multiple responses of interest, it is hard for AM designers to attain optimal part performance without a systematic approach. In this research, a data-driven framework is proposed to achieve the desired AM part performance and quality by predicting part properties and optimizing AM process parameters effectively and efficiently. The proposed framework encompasses efficient sampling of design space and establishing the initial experiment points. Based on established empirical data, surrogate models, are used to characterise influence of critical process parameters on responses on interest. Further, process maps can be generated for enhancing understanding on the influence of process parameters on responses of interests and AM process characteristics. Subsequently, multi-objective optimisation coupled with a multi criteria decision making technique is applied to determine an optimal design point, which maximises the identified responses of interest to meet the part functional requirements. A case study is used to validate the proposed framework for optimising an ULTEM™ 9085 fused filament fabrication part to meet its functional requirements of surface roughness and mechanical strength. From the case study, results indicate that the proposed approach is able to achieve good predictive results for responses of interest with a relatively small dataset. Further, process maps generated from the surrogate model provide a visual representation of the influence between responses of interest and critical process parameters for FFF process, which traditionally requires multiple investigations to arrive at similar conclusions.

Catheter-associated urinary tract infection reduction in critical care units: a bundled care model

Catheter-associated urinary tract infections (CAUTIs) represent approximately 9% of all hospital acquired infections, and approximately 65%–70% of CAUTIs are believed to be preventable. In the spring of 2013, Boston Medical Center (BMC) began an initiative to decrease CAUTI rates within its intensive care units (ICUs). A CAUTI taskforce convened and reviewed process maps and gap analyses. Based on Centers for Disease Control and Prevention (CDC) and Institute for Healthcare Improvement (IHI) guidelines, and delineated by the Healthcare Infection Control Practices Advisory Committee 2009 guidelines, all BMC ICUs sequentially implemented plan–do–study–act cycles based on which measures were most easily adaptable and believed to have the highest impact on CAUTI rates. Implementation of five care bundles spanned 5 years and included (1) processes for insertion and maintenance of foley catheters; (2) indications for indwelling foley catheters; (3) appropriate testing for CAUTIs; (4) alternatives to indwelling devices; and (5) sterilisation techniques. Daily rounds by unit nursing supervisors and inclusion of foley catheter necessity on daily ICU checklists held staff accountable on a daily basis. With these interventions, the total number of CAUTIs at BMC decreased from 53 in 2013 to 9 in 2017 (83% reduction) with a 33.8% reduction in indwelling foley catheter utilisation during the same time period. Adapted protocols showed success in decreasing the CAUTI rate and indwelling foley catheter usage in all of the BMC ICU’s. While all interventions had favourable and additive trends towards decreasing the CAUTI rate, the CAUTI awareness education, insertion and removal protocols and implementation of PureWick female incontinence devices had clear and significant effects on decreasing CAUTI rates. Our project provides a framework for improving HAIs using rapid cycle testing and U-chart data monitoring. Targeted education efforts and standardised checklists and protocols adapted sequentially are low-cost and high yield efforts that may decrease CAUTIs in ICU settings.

A novel approach to educate hospitalized cardiovascular disease patients about lifestyle and behavior modifications

Background Cardiovascular diseases (CVDs) are always considered by healthcare specialists for different reasons, including extensive prevalence, increased costs, chronicity, and high risk of death. The control of CVDs is highly influenced by behavior and lifestyle and it seems necessary to train special abilities about lifestyle and behavior modification to improve self-care skills for patients, and their caregivers. As a result, the development of effective training systems should be considered by healthcare specialists. Methods Hence, in this study, a framework for improving cardiovascular patients’ education processes is presented. Initially, an existing training system for cardiovascular patients is reviewed. Using field observations and targeted interviews with hospital experts, all components of its educating processes are identified, and their process maps are drawn up. After that, challenges in the training system are extracted with the aid of in-depth semi-structured interviews with experts. Due to the importance and different influence of the identified challenges, they are prioritized using a Multiple Criteria Decision-making (MCDM) method, and then their root causes were investigated. Finally, a novel framework is proposed and evaluated with hospital experts' help to improve the main challenges. Results The most important challenges included high nursing workload and shortage of time, lack of understanding of training concepts by patients, lack of attention to training, disruption of the training processes by the patients’ caregivers, and patient's weakness in understanding the standard language. In identifying the root causes, learner, educator, and educational tools are the most effective in the training process; therefore, the improvement scenarios were designed accordingly in the proposed framework. Conclusions Our study indicated that presenting a framework with applying different quantitative and qualitative methods has great potential to improve the processes of patient education for chronic diseases such as cardiovascular disease.

How has COVID-19 impacted the civil registration and vital statistics system in Loreto, Perú? Evidence using process mapping and qualitative analysis

ObjectivesAccurate civil registration and vital statistics (CRVS) systems are the primary data source to measure the impact of the COVID-19 pandemic on mortality. This study assesses how the pandemic impacted CRVS system processes in Loreto region of Peru, one of the worst affected countries globally.DesignQualitative study.SettingLoreto, a remote region, which had the highest reported mortality rate in Peru during the pandemic.ParticipantsSemistructured individual interviews and documentary analysis were conducted between September 2020 and May 2021 with 28 key informants from eight institutions involved in death certification. Key informants were identified using a purposive sampling strategy commencing at the Health Directorate of Loreto, and the snowball method was used where a participant suggested another organisation or person. Information from key informants was used to compare business process maps of the CRVS system before and during the pandemic.ResultsDuring early May 2020, there were seven times more registered deaths than in earlier years, but key informants believed this underestimated mortality by 20%–30%. During the pandemic, families had to interact with more institutions during the death certification process. Several issues disrupted death certification processes, including the burden of increased deaths, the Environmental Health Directorate often removing a body without the family’s express agreement, the creation of COVID-19 cemeteries where no death certificate was needed for burial, greater participation of funeral homes that often used outdated paper forms, and closure of civil registry offices. There was increased use of the online National Death System (SINADEF) but many users had problems with access.ConclusionsThe pandemic substantially disrupted CRVS processes in Loreto, making death certification more difficult, placing greater burden on the family and leading to more participation from unregulated organisations such as funeral homes or cemeteries. These disruptions were impacted by limitations of the CRVS system’s processes before the pandemic.

A time-driven activity-based costing approach for identifying variability in costs of childbirth between and within types of delivery

Background Recently, time-driven activity-based costing (TDABC) is put forward as an alternative, more accurate costing method to calculate the cost of a medical treatment because it allows the assignment of costs directly to patients. The objective of this paper is the application of a time-driven activity-based method in order to estimate the cost of childbirth at a maternal department. Moreover, this study shows how this costing method can be used to outline how childbirth costs vary according to considered patient and disease characteristics. Through the use of process mapping, TDABC allows to exactly identify which activities and corresponding resources are impacted by these characteristics, leading to a more detailed understanding of childbirth cost. Methods A prospective cohort study design is performed in a maternity department. Process maps were developed for two types of childbirth, vaginal delivery (VD) and caesarean section (CS). Costs were obtained from the financial department and capacity cost rates were calculated accordingly. Results Overall, the cost of childbirth equals €1894,12 and is mainly driven by personnel costs (89,0%). Monitoring after birth is the most expensive activity on the pathway, costing €1149,70. Significant cost variations between type of delivery were found, with VD costing €1808,66 compared to €2463,98 for a CS. Prolonged clinical visit (+ 33,3 min) and monitoring (+ 775,2 min) in CS were the main contributors to this cost difference. Within each delivery type, age, parity, number of gestation weeks and education attainment were found to drive cost variations. In particular, for VD an age > 25 years, nulliparous, gestation weeks > 40 weeks and higher education attainment were associated with higher costs. Similar results were found within CS for age, parity and number of gestation weeks. Conclusions TDABC is a valuable approach to measure and understand the variability in costs of childbirth and its associated drivers over the full care cycle. Accordingly, these findings can inform health care providers, managers and regulators on process improvements and cost containment initiatives.

Requirements and process analysis for ports and waterways openBIM ISO standards development

Purpose Defining building information modelling (BIM) standards for the infrastructure domain is a central issue to the successful implementation of BIM in civil engineering domains. To this end, this paper aims to present a requirements and process analysis for the ports and waterways domain to address the lack of BIM standards development, using the information delivery manual (IDM) approach and the ethos of openBIM standards. Design/methodology/approach This research uses the IDM approach. This involves the definition of use cases, process maps, exchange scenarios and subsequent exchange requirements. All these developments were sourced and validated by a series of international industry consultations. Findings The paper identifies 30 domain relevant use cases collated from existing sources and new cases. An overview and detailed ports and waterways process map (defining actors, activities and data exchanges). The process maps highlighted 38 exchange scenarios between various activities. Various exchange requirements were defined and are discussed in the context of the required information exchange model and the extensions required to fulfil the needs of the domain. The analysis provides the core information for the next steps of development for a substantial extension to the Industry Foundation Classes and the supporting data dictionary standards. Research limitations/implications Because of the international scope of the research, the outcomes can be applied by any stakeholders in the domain of ports and waterways. Therefore, some variation is expected at a national and organizational level. This research has the potential to accelerate the adoption of openBIM standards within the ports and waterways domain leading to increases in efficiency, collaborative working. Originality/value This paper reviews the requirements of an identified gap in the provision of openBIM standards relevant and applicable to the domain of ports and waterways.

The Process of Designing the Post-Mining Land Reclamation Investment Using Process Maps. Case Study

In the present day, Europe is moving away from fossil fuels. This is due to the need to reduce CO2 emissions released to the atmosphere. As a result of this process, especially in Upper Silesia, Poland, there will be more areas requiring measures after the closing of mines. Reclamation of post-mining areas is a very difficult task, as there is no universal method of planning the recultivation. During mining operations, we observed many forms of environmental degradation. The article presents the use of one of the newer and increasingly used modeling methods in recultivation processes—the process map. Analysis of the investment is presented, based on the example of the recultivation of a pit after closing the hard coal mine. The main purpose of the analysis was to shorten the designing time of the construction and realization process. Two maps of processes were made, and a map of the existing and desired processes, as well as the benefits from this, are shown. This article presents only one of the stages of the recultivation process—the investment project process—, treating it as an example for optimizing the entire project related to recultivation.

Multi-scale simulation approach for identifying optimal parameters for fabrication ofhigh-density Inconel 718 parts using selective laser melting

Purpose Depending on an experimental approach to find optimal parameters for producing fully dense (relative density > 99%) Inconel 718 (IN718) components in the selective laser melting (SLM) process is expensive and offers no guarantee of success. Accordingly, this study aims to propose a multi-scale simulation framework to guide the choice of processing parameters in a more pragmatic manner. Design/methodology/approach In the proposed approach, a powder layer, ray tracing and heat transfer simulation models are used to calculate the melt pool dimensions and evaporation volume corresponding to a small number of laser power and scanning speed conditions within the input design space. A layer-heating model is then used to determine the inter-layer idle time required to maximize the temperature convergence rate of the solidified layer beneath the power bed. The simulation results are used to train surrogate models to construct SLM process maps for 3,600 pairs of the laser power and scanning speed within the input design space given three different values of the underlying solidified layer temperature (i.e., 353 K, 673 K and 873 K). The ideal selection of laser power and scanning speed of each process map is chosen based on four quality-related criteria listed as follows: without the appearance of key-hole melting; an evaporation volume less than the volume of the d90 powder particles; ensuring the stability of single scan tracks; and avoiding a weak contact between the melt pool and substrate. Finally, the optimal laser power and scanning speed parameters for the SLM process are determined by superimposing the optimal regions of the individual process maps. Findings The feasibility of the proposed approach is demonstrated by fabricating IN718 test specimens using the optimal processing conditions identified by the simulation framework. It is shown that the maximum density of the fabricated parts is 99.94%, while the average density is 99.88% and the standard deviation is less than 0.05%. Originality/value The present study proposed a multi-scale simulation model which can efficiently predict the optimal processing conditions for producing fully dense components in the SLM process. If the geometry of the three-dimensional printed part is changed or the machine and powder material is altered, users can use the proposed method for predicting the processing conditions that can produce the high-density part.

Application of Deep Transfer Learning and Uncertainty Quantification for Process Identification in Powder Bed Fusion

Accurate identification and modeling of process maps in additive manufacturing remains a pertinent challenge. To ensure high quality and reliability of the finished product researchers rely on models that entail the physics of the process as a computer code or conduct laboratory experiments which are expensive and oftentimes demands significant logistic and overheads. Physics based computational modeling has shown promise in alleviating the aforementioned challenge, albeit with limitations like physical approximations, model-form uncertainty, and limited experimental data. This calls for modeling methods that can combine limited experimental and simulation data in a computationally efficient manner, in order to achieve the desired properties in the manufactured parts. In this paper, we focus on demonstrating the impact of probabilistic modeling and uncertainty quantification on powder-bed fusion additive manufacturing by focusing on the following three milieu: a) accelerating the parameter development processes associated with laser powder bed fusion additive manufacturing process of metals, b) quantifying uncertainty and identifying missing physical correlations in the computational model, and c) transferring learned process maps from a source to a target process. These tasks demonstrate the application of multi-fidelity modeling, global sensitivity analysis, intelligent design of experiments and deep transfer learning for a meso-scale meltpool model of the additive manufacturing process.

PoreAnalyzer—An Open-Source Framework for the Analysis and Classification of Defects in Additive Manufacturing

Additive manufacturing processes offer high geometric flexibility and allow the use of new alloy concepts due to high cooling rates. For each new material, parameter studies have to be performed to find process parameters that minimize microstructural defects such as pores or cracks. In this paper, we present a system developed in Python for accelerated image analysis of optical microscopy images. Batch processing can be used to quickly analyze large image sets with respect to pore size distribution, defect type, contribution of defect type to total porosity, and shape accuracy of printed samples. The open-source software is independent of the microscope used and is freely available for use. This framework allows us to perform such an analysis on a circular area with a diameter of 5 mm within 10 s, allowing detailed process maps to be obtained for new materials within minutes after preparation.