Optimizing an Algorithm for Data Mining a Design Repository to Automate Functional Modeling

2020 ◽  
Author(s):  
Alex Mikes ◽  
Katherine Edmonds ◽  
Robert B. Stone ◽  
Bryony DuPont
Keyword(s):  
Data Mining ◽  
Functional Model ◽  
Prediction Algorithm ◽  
Optimum Size ◽  
Functional Modeling ◽  
Training Set ◽  
Specific Focus ◽  
Average Accuracy ◽  
Functional Models

Abstract The purpose of this research is to find the optimum values for threshold variables used in a data mining and prediction algorithm. We also minimize and stratify a training set to find the optimum size based on how well it represents the whole dataset. Our specific focus is automating functional models, but the method can be applied to any dataset with a similar structure. We iterate through different values for two of the threshold variables in this process and cross-validate to calculate the average accuracy and find the optimum values for each variable. We optimize the training set by reducing the size by 78% and stratifying the data, whereby we achieve an accuracy that is 96% as good as the whole training set and takes 50% less time. These optimum values can be used to better predict the functions and flows of any future product based on its constituent components, which can be used to generate a complete functional model.

Experimental Studies Assessing the Repeatability of a Functional Modeling Derivation Method

2003 ◽  
Vol 125 (4) ◽  
pp. 682-693 ◽  
Author(s):  
Mark A. Kurfman ◽  
Michael E. Stock ◽  
Robert B. Stone ◽  
Jagan Rajan ◽  
Kristin L. Wood
Keyword(s):  
Design Research ◽  
Functional Model ◽  
Experimental Studies ◽  
Testing Procedure ◽  
Research Literature ◽  
Functional Modeling ◽  
Skill Levels ◽  
Functional Models ◽  
Model Derivation ◽  
Derivation Method

This paper presents the results of research attempts to substantiate repeatability and uniqueness claims of a functional model derivation method following a hypothesis generation and testing procedure outlined in design research literature. Three experiments are constructed and carried out with a participant pool that possesses a range of engineering design skill levels. The experiments test the utility of a functional model derivation method to produce repeatable functional models for a given product among different designers. In addition to this, uniqueness of the functional models produced by the participants is examined. Results indicate the method enhances repeatability and leads designers toward a unique functional model of a product. Shortcomings of the method and opportunities for improvement are also identified.

Mining Functional Model Graphs to Find Product Design Heuristics With Inclusive Design Illustration

2013 ◽  
Vol 13 (4) ◽  
Author(s):  
Shraddha Sangelkar ◽  
Daniel A. McAdams
Keyword(s):  
Data Mining ◽  
Functional Model ◽  
Inclusive Design ◽  
Generation Process ◽  
Graph Visualization ◽  
Concept Generation ◽  
Product Representation ◽  
Graph Data ◽  
Design Heuristics ◽  
Functional Models

Engineering design heuristics offer the potential to improve the design process and resultant designs. Currently, heuristics are empirically derived by experts. The goal of this paper is to automate the heuristics generation process. Functional modeling, a well-established product representation framework, is applied in this research to abstract the intended functionality of a product. Statistically significant heuristics, extracted from a database of functional models, serve as design suggestions or guidelines for concept generation. The heuristics can further be applied to automate portions of the concept generation process. Prior research efforts in automated concept generation rely heavily on the design repository. The repository needs to be appended for broader categories of design problems, and, at the same time, a tool for quick analysis of the expanded repository is required. An automated heuristic extraction process has the capability to efficiently mine the updated repositories and find new heuristics for design practice. A key objective of this research is to develop design heuristics applicable in the diverse and challenging domain of inclusive design. The research applies graph theory for mathematical representation of the functional model, graph visualization for comprehending graphs, and graph data mining to extract heuristics. The results show that the graphical representation of functional models along with graph visualization allows quick updates to the design repository. In addition, we show that graph data mining has the capability to efficiently search for new design heuristics from the updated repository.

Functional Modeling Experimental Studies

2001 ◽  
Author(s):  
Mark A. Kurfman ◽  
Robert B. Stone ◽  
Jagan R. Rajan ◽  
Kristin L. Wood
Keyword(s):  
Engineering Design ◽  
Functional Model ◽  
Experimental Studies ◽  
Design Engineering ◽  
Functional Modeling ◽  
Skill Levels ◽  
Functional Models ◽  
New Methodologies ◽  
Model Derivation ◽  
Derivation Method

Abstract As more design methodologies are researched and developed, the question arises as to whether these new methodologies are actually advancing the field of engineering design or instead cluttering the field with more theories. There is a critical need to test new methodologies for their contribution to the field of design engineering. This paper presents the results of research attempts to substantiate repeatability claims of the functional model derivation method. Three experiments are constructed and carried out with a participant pool that possesses a range of engineering design skill levels. The experiments test the utility of the functional model derivation method to produce repeatable functional models for a given product among different designers. Results indicate the method is largely successful and identify its key strengths as well as opportunities for improvement.

Question-by-Question Interrater Analysis and Suggestions for Improvements of a Functional Model Scoring Rubric

2018 ◽  
Author(s):  
Alexander R. Murphy ◽  
Jacob T. Nelson ◽  
Matt R. Bohm ◽  
Robert L. Nagel ◽  
Julie S. Linsey
Keyword(s):  
Interrater Reliability ◽  
Functional Model ◽  
System Function ◽  
Functional Modeling ◽  
Component Structure ◽  
Functional Models ◽  
Ongoing Development ◽  
Evaluation Metric ◽  
Vague Language ◽  
Functional Components

Functional modeling is a tool used for system abstraction. By divorcing system function from component structure, functional modeling allows designers to more easily identify design opportunities and compartmentalize product functions, which can lead to innovation during the ideation process. In this paper, we examine the reliability of a rubric used to evaluate student-generated functional models by comparing interrater reliabilities on a question-by-question basis from a previous study where an examination of the reliability of each question was not assessed. We then suggest changes to the rubric in order to improve the rubric’s overall interrater reliability as well as its question-by-question interrater reliability. These rubric alterations include clarification of vague language, inclusion of examples and counter examples, and a procedure for handling nonexistent functional components as opposed to incorrect or “nonsensical” functional components. This work is in contribution to the ongoing development of this functional modeling rubric as an education instrument. As functional modeling becomes more widely accepted in the design community and in engineering curricula, it is important to have a validated evaluation metric with which to assess student-generated functional models.

A Theory for the Development of Conceptual Functional Models for Automation of Manual Processes

2007 ◽  
Author(s):  
Robert L. Nagel ◽  
Robert B. Stone ◽  
Daniel A. McAdams
Keyword(s):  
Conceptual Design ◽  
Functional Model ◽  
Process Models ◽  
Functional Modeling ◽  
Functional Basis ◽  
Functional Models ◽  
Electromechanical Products ◽  
The Creation

Conceptual design is a vital stage in the development of any product, and its importance only increases with the complexity of a design. Functional modeling with the Functional Basis provides a framework for the conceptual design of electromechanical products. This framework is just as applicable to the conceptual design of automated solutions where an engineered product with components spanning multiple engineering domains is designed to replace or aid a human and his or her tools in a human-centric process. This paper presents research toward the simplification of the generation of conceptual functional models for automation solutions. The presented methodology involves the creation of functional and process models to fully explore existing human operated tasks for potential automation. Generated functional and process models are strategically combined to create a new conceptual functional model for an automation solution to potentially automate the human-centric task. The presented methodology is applied to the generation of a functional model for a conceptual automation solution. Then conceptual automation solutions generated through the presented methodology are compared to existing automation solutions to demonstrate the effectiveness of the presented methodology.

Improving Students' Functional Modeling Skills: A Modeling Approach and a Scoring Rubric

2015 ◽  
Vol 137 (5) ◽  
Author(s):  
Robert L. Nagel ◽  
Matt R. Bohm ◽  
Julie S. Linsey ◽  
Marie K. Riggs
Keyword(s):  
Engineering Design ◽  
Functional Model ◽  
Concept Generation ◽  
Functional Modeling ◽  
Generate Functional ◽  
Algorithmic Approach ◽  
Functional Models ◽  
Grammar Rules ◽  
Students First ◽  
Scoring Rubric

An engineering design curriculum that introduces functional modeling methods is believed to enhance the ability to abstract complex systems, assist during the concept generation phase of design, and reduce design fixation. To that end, a variety of techniques for considering function during design have been proposed in the literature, yet there are a lack of validated approaches for teaching students to generate functional models and no reliable method for the assessment of functional models. This paper presents a study investigating students' ability to generate functional models during a homework assignment; the study includes three different treatment conditions: (1) students who receive only a lecture on functional modeling, (2) students who receive a lecture on functional modeling as well as a step-by-step example, and (3) students who receive a lecture, a step-by-step example, and an algorithmic approach with grammar rules. The experiment was conducted in a cornerstone, undergraduate engineering design course, and consequently, was the students' first exposure to functional modeling. To assess student generated functional models across all three conditions, an 18 question functional model scoring rubric was developed based on flow-based functional modeling standards. Use of the rubric to assess the student generated functional models resulted in high inter-rater agreement for total score. Results show that students receiving the step-by-step example perform as well as students receiving the step-by-step example and an algorithmic approach with grammar rules; both groups perform better than the lecture-only group.

scholarly journals Designing municipal library functional model

2016 ◽  
pp. 5-15
Author(s):  
Inna Doronina ◽  
Svetlana Londar
Keyword(s):  
Functional Model ◽  
Information Services ◽  
Belgorod Oblast ◽  
Functional Modeling ◽  
Functional Models ◽  
Efficient Information

The conclusions on using functional modeling results for library processes modernization are presented. The authors prove that such functional models make the basis for efficient information services. The study was accomplished at Valuy district Intersettlement Library, Belgorod Oblast, Russia.

A Weighted Confidence Metric to Improve Automated Functional Modeling

2020 ◽  
Author(s):  
Katherine Edmonds ◽  
Alex Mikes ◽  
Bryony DuPont ◽  
Robert B. Stone
Keyword(s):  
Linear Functional ◽  
Individual Component ◽  
Functional Model ◽  
Functional Modeling ◽  
Additional Information ◽  
Functional Models ◽  
Wide Range ◽  
Ultimate Objective ◽  
And Function ◽  
Component Function

Abstract Expanding on previous work of automating functional modeling, we have developed a more informed automation approach by assigning a weighted confidence metric to the wide variety of data in a design repository. Our work focuses on automating what we call linear functional chains, which are a component-based section of a full functional model. We mine the Design Repository to find correlations between component and function and flow. The automation algorithm we developed organizes these connections by component-function-flow frequency (CFF frequency), thus allowing the creation of linear functional chains. In previous work, we found that CFF frequency is the best metric in formulating the linear functional chain for an individual component; however, we found that this metric did not account for prevalence and consistency in the Design Repository data. To better understand our data, we developed a new metric, which we refer to as weighted confidence, to provide insight on the fidelity of the data, calculated by taking the harmonic mean of two metrics we extracted from our data, prevalence, and consistency. This method could be applied to any dataset with a wide range of individual occurrences. The contribution of this research is not to replace CFF frequency as a method of finding the most likely component-function-flow correlations but to improve the reliability of the automation results by providing additional information from the weighted confidence metric. Improving these automation results, allows us to further our ultimate objective of this research, which is to enable designers to automatically generate functional models for a product given constituent components.

PREDIKSI KUALITAS AIR SUNGAI CILIWUNG DENGAN MENGGUNAKAN ALGORITMA POHON KEPUTUSAN

2021 ◽  
Vol 12 (2) ◽  
Author(s):  
Mohammad Haekal ◽  
Henki Bayu Seta ◽  
Mayanda Mega Santoni
Keyword(s):  
Data Mining ◽  
Decision Tree ◽  
Cross Validation ◽  
Online Monitoring ◽  
Training Set ◽  
Microsoft Excel ◽  
Test Set

Untuk memprediksi kualitas air sungai Ciliwung, telah dilakukan pengolahan data-data hasil pemantauan secara Online Monitoring dengan menggunakan Metode Data Mining. Pada metode ini, pertama-tama data-data hasil pemantauan dibuat dalam bentuk tabel Microsoft Excel, kemudian diolah menjadi bentuk Pohon Keputusan yang disebut Algoritma Pohon Keputusan (Decision Tree) mengunakan aplikasi WEKA. Metode Pohon Keputusan dipilih karena lebih sederhana, mudah dipahami dan mempunyai tingkat akurasi yang sangat tinggi. Jumlah data hasil pemantauan kualitas air sungai Ciliwung yang diolah sebanyak 5.476 data. Hasil klarifikasi dengan Pohon Keputusan, dari 5.476 data ini diperoleh jumlah data yang mengindikasikan sungai Ciliwung Tidak Tercemar sebanyak 1.059 data atau sebesar 19,3242%, dan yang mengindikasikan Tercemar sebanyak 4.417 data atau 80,6758%. Selanjutnya data-data hasil pemantauan ini dievaluasi menggunakan 4 Opsi Tes (Test Option) yaitu dengan Use Training Set, Supplied Test Set, Cross-Validation folds 10, dan Percentage Split 66%. Hasil evaluasi dengan 4 opsi tes yang digunakan ini, semuanya menunjukkan tingkat akurasi yang sangat tinggi, yaitu diatas 99%. Dari data-data hasil peneltian ini dapat diprediksi bahwa sungai Ciliwung terindikasi sebagai sungai tercemar bila mereferensi kepada Peraturan Pemerintah Republik Indonesia nomor 82 tahun 2001 dan diketahui pula bahwa penggunaan aplikasi WEKA dengan Algoritma Pohon Keputusan untuk mengolah data-data hasil pemantauan dengan mengambil tiga parameter (pH, DO dan Nitrat) adalah sangat akuran dan tepat. Kata Kunci : Kualitas air sungai, Data Mining, Algoritma Pohon Keputusan, Aplikasi WEKA.

Sign in / Sign up

Export Citation Format

Share Document