Validating a Tool for Predicting Problem-Specific Optimized Team Characteristics

2017 ◽  
Author(s):  
Christopher McComb ◽  
Jonathan Cagan ◽  
Kenneth Kotovsky
Keyword(s):  
Design Problem ◽  
State Of The Art ◽  
Response Surfaces ◽  
Problem Instance ◽  
Configuration Design ◽  
Team Characteristics ◽  
Optimal Approach ◽  
Cognitive Study ◽  
Superior Approach ◽  
Novel Design

The performance of a team is highly dependent on how the team is structured, how individuals in the team communicate with one another, and the properties exhibited by the problem being solved. It is generally assumed that teams are a superior approach in problem-solving and design. However, this work shows that for a configuration design problem of moderate size, the optimal approach for a homogenous team is in fact for members of the team to work independently, with the best solution from the individuals chosen at the end. Moreover, this work demonstrates that this surprising strategy can be predicted from knowledge of the problem’s properties through a computationally-derived set of response surfaces. First, a novel design problem is defined that requires solvers to create a system of internet-connected products to maintain the temperature within a home. Next, the characteristics of this new design problem are measured, and a computationally-derived response surface yields the untraditional prediction that teams should not interact while solving the problem. Finally, this prediction is tested and shown correct through a cognitive study. This work makes two contributions to the state of the art. First, it provides verification of a methodology that allows optimal team characteristics to be predicted based on knowledge of a design problem. Second, it demonstrates an additional problem instance for which interacting teams are inferior to nominal teams (adding to a growing literature to that effect).

scholarly journals Optimizing Design Teams Based on Problem Properties: Computational Team Simulations and an Applied Empirical Test

2018 ◽  
Author(s):  
Christopher McComb ◽  
Kenneth Kotovsky ◽  
Jonathan Cagan
Keyword(s):  
Empirical Test ◽  
Configuration Design ◽  
Design Teams ◽  
Design Problems ◽  
Team Members ◽  
Optimizing Design ◽  
Team Characteristics ◽  
Cognitive Study ◽  
The Right ◽  
Predictive Theory

The performance of a team with the right characteristics can exceed the mere sum of the constituent members’ individual efforts. However, a team having the wrong characteristics may perform more poorly than the sum of its individuals. Therefore, it is vital that teams are assembled and managed properly in order to maximize performance. This work examines how the properties of configuration design problems can be leveraged to select the best values for team characteristics (specifically team size and interaction frequency). A computational model of design teams which has been shown to effectively emulate human team behavior is employed to pinpoint optimized team characteristics for solving a variety of configuration design problems. These configuration design problems are characterized with respect to the local and global structure of the design space, the alignment between objectives, and the resources allotted for solving the problem. Regression analysis is then used to create equations for predicting optimized values for team characteristics based on problem properties. These equations achieve moderate to high accuracy, making it possible to design teams based on those problem properties. Further analysis reveals hypotheses about how the problem properties can influence a team’s search for solutions. This work also conducts a cognitive study on a different problem to test the predictive equations. For a configuration problem of moderate size, the model predicts that zero interaction between team members should lead to the best outcome. A cognitive study of human teams verifies this surprising prediction, offering partial validation of the predictive theory.

scholarly journals Optimizing Design Teams Based on Problem Properties: Computational Team Simulations and an Applied Empirical Test

2017 ◽  
Vol 139 (4) ◽  
Author(s):  
Christopher McComb ◽  
Jonathan Cagan ◽  
Kenneth Kotovsky
Keyword(s):  
Empirical Test ◽  
Configuration Design ◽  
Design Teams ◽  
Design Problems ◽  
Team Members ◽  
Optimizing Design ◽  
Team Characteristics ◽  
Cognitive Study ◽  
The Right ◽  
Predictive Theory

The performance of a team with the right characteristics can exceed the mere sum of the constituent members' individual efforts. However, a team having the wrong characteristics may perform more poorly than the sum of its individuals. Therefore, it is vital that teams are assembled and managed properly in order to maximize performance. This work examines how the properties of configuration design problems can be leveraged to select the best values for team characteristics (specifically team size and interaction frequency). A computational model of design teams which has been shown to effectively emulate human team behavior is employed to pinpoint optimized team characteristics for solving a variety of configuration design problems. These configuration design problems are characterized with respect to the local and global structure of the design space, the alignment between objectives, and the resources allotted for solving the problem. Regression analysis is then used to create equations for predicting optimized values for team characteristics based on problem properties. These equations achieve moderate to high accuracy, making it possible to design teams based on those problem properties. Further analysis reveals hypotheses about how the problem properties can influence a team's search for solutions. This work also conducts a cognitive study on a different problem to test the predictive equations. For a configuration problem of moderate size, the model predicts that zero interaction between team members should lead to the best outcome. A cognitive study of human teams verifies this surprising prediction, offering partial validation of the predictive theory.

scholarly journals Translucent Answer Predictions in Multi-Hop Reading Comprehension

2020 ◽  
Vol 34 (05) ◽  
pp. 7700-7707
Author(s):  
G P Shrivatsa Bhargav ◽  
Michael Glass ◽  
Dinesh Garg ◽  
Shirish Shevade ◽  
Saswati Dana ◽  
...  
Keyword(s):  
Reading Comprehension ◽  
Question Answering ◽  
State Of The Art ◽  
Local Context ◽  
The Novel ◽  
Loose Coupling ◽  
Loosely Coupled ◽  
Neural Architecture ◽  
Coupled Networks ◽  
Novel Design

Research on the task of Reading Comprehension style Question Answering (RCQA) has gained momentum in recent years due to the emergence of human annotated datasets and associated leaderboards, for example CoQA, HotpotQA, SQuAD, TriviaQA, etc. While state-of-the-art has advanced considerably, there is still ample opportunity to advance it further on some important variants of the RCQA task. In this paper, we propose a novel deep neural architecture, called TAP (Translucent Answer Prediction), to identify answers and evidence (in the form of supporting facts) in an RCQA task requiring multi-hop reasoning. TAP comprises two loosely coupled networks – Local and Global Interaction eXtractor (LoGIX) and Answer Predictor (AP). LoGIX predicts supporting facts, whereas AP consumes these predicted supporting facts to predict the answer span. The novel design of LoGIX is inspired by two key design desiderata – local context and global interaction– that we identified by analyzing examples of multi-hop RCQA task. The loose coupling between LoGIX and the AP reveals the set of sentences used by the AP in predicting an answer. Therefore, answer predictions of TAP can be interpreted in a translucent manner. TAP offers state-of-the-art performance on the HotpotQA (Yang et al. 2018) dataset – an apt dataset for multi-hop RCQA task – as it occupies Rank-1 on its leaderboard (https://hotpotqa.github.io/) at the time of submission.

Decoupling spatial and temporal processes for clinical analyzers

1995 ◽  
Vol 41 (9) ◽  
pp. 1398-1402
Author(s):  
J Mazza ◽  
M Huber ◽  
S Frye
Keyword(s):  
Number Theory ◽  
Design Problem ◽  
State Of The Art ◽  
Spatial Arrangement ◽  
Sample Volume ◽  
Complex Task ◽  
Clinical Testing ◽  
Time And Space ◽  
Physical Processing ◽  
Temporal And Spatial

Abstract The separation of time and space in processing a sample greatly simplifies the design of automation for clinical testing. The efficient spatial arrangement of analytical units and sample manipulators has become a more complex task because of the degree of automation required on today's state-of-the-art analyzer. Minimization of sample volume and the reduction of overall analyzer size further complicate the design problem. We report the development of a proprietary method of decoupling the temporal and spatial elements required for analysis of samples. This process is based on number theory and can be used to optimize the distance between the physical processing stations while allowing these same stations to operate on samples over a substantial range of times. The technique is versatile and can also be used when it is desirable to sequentially move groups of items from location to location.

scholarly journals CASE STUDY IN SYSTEMATIC DESIGN ENGINEERING – STAGE BARGE GANGWAY

2013 ◽  
Author(s):  
W. Ernst Eder
Keyword(s):  
Engineering Design ◽  
Design Process ◽  
Design Problem ◽  
Design Science ◽  
Process Engineering ◽  
Design Engineering ◽  
Learning Design ◽  
Systematic Design ◽  
Novel Design

Students learning design engineering at times need a good example of procedure for novel design engineering. The systematic heuristic-strategic use of a theory to guide the design process – Engineering Design Science – and the methodical design process followed in this case study is only necessary in limited situations. The full procedure should be learned, such that the studentcan select appropriate parts for other applications.This case example is presented to show application of the recommended method, and the expected scope of the output, with emphasis on the stages of conceptualizing. The case follows a novel design problem of a gangway for ship-shore transfer for the Caravan Stage Barge.

scholarly journals Automated Algorithm Selection: Survey and Perspectives

2019 ◽  
Vol 27 (1) ◽  
pp. 3-45 ◽  
Author(s):  
Pascal Kerschke ◽  
Holger H. Hoos ◽  
Frank Neumann ◽  
Heike Trautmann
Keyword(s):  
State Of The Art ◽  
Combinatorial Problems ◽  
The State ◽  
Problem Instance ◽  
Ai Planning ◽  
Algorithm Selection ◽  
Automated Algorithm ◽  
Continuous Problems ◽  
Algorithm Configuration ◽  
The One

It has long been observed that for practically any computational problem that has been intensely studied, different instances are best solved using different algorithms. This is particularly pronounced for computationally hard problems, where in most cases, no single algorithm defines the state of the art; instead, there is a set of algorithms with complementary strengths. This performance complementarity can be exploited in various ways, one of which is based on the idea of selecting, from a set of given algorithms, for each problem instance to be solved the one expected to perform best. The task of automatically selecting an algorithm from a given set is known as the per-instance algorithm selection problem and has been intensely studied over the past 15 years, leading to major improvements in the state of the art in solving a growing number of discrete combinatorial problems, including propositional satisfiability and AI planning. Per-instance algorithm selection also shows much promise for boosting performance in solving continuous and mixed discrete/continuous optimisation problems. This survey provides an overview of research in automated algorithm selection, ranging from early and seminal works to recent and promising application areas. Different from earlier work, it covers applications to discrete and continuous problems, and discusses algorithm selection in context with conceptually related approaches, such as algorithm configuration, scheduling, or portfolio selection. Since informative and cheaply computable problem instance features provide the basis for effective per-instance algorithm selection systems, we also provide an overview of such features for discrete and continuous problems. Finally, we provide perspectives on future work in the area and discuss a number of open research challenges.

Increasing Innovation: A Trilogy of Experiments Towards a Design-by-Analogy Method

2007 ◽  
Author(s):  
J. S. Linsey ◽  
J. P. Laux ◽  
E. Clauss ◽  
K. L. Wood ◽  
A. B. Markman
Keyword(s):  
Design Problem ◽  
Domain Specific Language ◽  
General Representation ◽  
Design Problems ◽  
Analogy Method ◽  
Domain Specific ◽  
Design By Analogy ◽  
Product Description ◽  
Novel Design ◽  
Analogous Product

Design by analogy is a noted approach for conceptual design. This paper seeks to develop a robust design-by-analogy method. This endeavor is sought through a series of three experiments focusing on understanding the influence of representation on the design-by-analogy process. The first two experiments evaluate the effects of analogous product description—presented in either domain-general or domain-specific language—on a designer’s ability to later use the product to solve a novel design problem. Six different design problems with corresponding analogous products are evaluated. The third experiment in the series uses a factorial design to explore the effects of the representation (domain specific or general sentinel descriptions) for both the design problem and the analogous product on the designer’s ability to develop solutions to novel design problems. Results show that a more general representation of the analogous products facilitates later use for a novel design problem. The highest rates of success occur when design problems are presented in domain specific representations and the analogous product is in a domain general representation. Other insights for the development of design by analogy methods and tools are also discussed.

scholarly journals Data on the Configuration Design of Internet-Connected Home Cooling Systems by Engineering Students

2018 ◽  
Author(s):  
Christopher McComb ◽  
Jonathan Cagan ◽  
Kenneth Kotovsky
Keyword(s):  
Design Problem ◽  
Design Pattern ◽  
Engineering Students ◽  
Configuration Design ◽  
Cooling Systems ◽  
Analysis And Design ◽  
Cost Constraints ◽  
Total Data ◽  
Computer Based ◽  
Study Participants

This experiment was carried out to record the step-by-step actions that humans take in solving a configuration design problem, either in small teams or individually. Specifically, study participants were tasked with configuring an internet-connected system of products to maintain temperature within a home, subject to cost constraints. Every participant was given access to a computer-based design interface that allowed them to construct and assess solutions. The interface was also used to record the data that is presented here. In total, data was collected for 68 participants, and each participant was allowed to perform 50 design actions in solving the configuration design problem. Major results based on the data presented here have been reported separately, including initial behavioral analysis and design pattern assessments via Markovian modeling.

Sign in / Sign up

Export Citation Format

Share Document