TEK 161 Exam | Questions and Answers (Complete Solutions)

TEK 161 Exam | Questions and Answers (Complete Solutions) Which of the following QFD practices best describes how it supports system thinking in an organization? A) It uses a set of linked houses of quality which are built upon maximizing the number of people in an organization. This is to maximize the number of different perspectives on formulating the customer needs. B) It uses a set of linked matrices of which one matrix' output representing a function in the organization is used as the input of the subsequent matrix representing another function in the organization. C) It uses a set of linked HoQs which are built upon minimizing the number of different functions involved in an organization. This is to minimize the unnecessary cost from various functional groups. D) It uses a set of linked matrices of which one matrix' output representing a function in the organization is independent of the input of the subsequent matrix representing another function in the organization. Which of the following statements best describes how the Arrow's impossibility theorem can be useful for Quality Function Deployment practitioners? A) It is impossible to assume a causal arrow between a demanded quality (the What) and a quality characteristic (the How). B) It is impossible to uncover all customer needs. C) It is impossible to satisfy several groups of customers. D) It is possible to satisfy several groups of customers. In the context of redesigning a product (e.g. a water bottle), which of the following questions is the most useful for uncovering customer needs according to Ulwick and Bettencourt (2008)? A) Start with why, ask "why do you buy a water bottle?" B) Focus on pain points, ask "what makes using a water bottle inconvenient?" C) Be straightforward, ask "what do you need from a water bottle?" D) Have an improvement-minded attitude, ask "how would you improve the design of this water bottle?" Within the context of understanding customer's needs and/or wants, there are four types of pleasure in products (Jordan, 2002), which of the following types of pleasure is best illustrated by the diamond's advertisement example shown in the class? A) psycho-pleasure B) physio-pleasure C) socio-pleasure D) ideo-pleasure Within the context of design thinking, if one aims to improve quality by redesigning a certain service experience where there is little knowledge of customer needs, which of the following activities is the best to start with? A) Carry out a customer satisfaction survey to know what worked well and what did not work well B) Perform the analytics, such as text mining on customer's complaints report and online reviews, then identify the problematic areas C) Carry out a customer satisfaction survey to know what did not work well D) Conduct some in-depth interviews focusing on the job to be done for the customers Within the context of design thinking as suggested by IDEO, which of the following statements best describes how one may get a good idea during the concepting phase? A) Separate the vital few ideas from the trivial many. B) Start with a few logical and feasible ideas. C) Collect ideas from heavy users through in-depth interviews focusing on the job to be done. D) Brainstorm as many ideas as possible regardless of their quality. One of the mechanisms in resourceful sensemaking (Beverland et al., 2016) is called exposure. Which of the following statements best describes the aim of exposure? A) to primarily facilitate informal interactions B) to primarily facilitate formal interactions C) to reveal interpretative schemes of other departments D) to generate more communication between departments Which of the following pairs of functional and dysfunctional questions is most useful for classifying a new service feature into Kano model's category? A) "How would you feel if you were provided with a voice-operated multimedia system that works well in your room?" and "How would you feel if you were provided with a voice operated multimedia system that works poorly in your room?" B) "How would you feel if you were provided with a voice-operated multimedia system in your room?" and "How would you feel if you were not provided with a voice-operated multimedia system in your room?" C) "How would you feel if you were provided with an easy-to-control voice-operated multimedia system in your room?" and "How would you feel if you were provided with a difficult-to control voice-operated multimedia system in your room?" D) "How would you feel if you were provided with a voice-operated multimedia system that works in your room?" and "How would you feel if you were provided with a voice-operated multimedia system that does not work in your room?" Which of the following statements best describes what it means to assume a neutral reference distribution when performing exploratory data analysis? A) To assume a non-normal distribution as a reference. B) To assume a normal distribution as a reference. C) To establish a reference to which the data will be compared. D) To establish a reference to which the theory will be compared. If you have a budget for making 16 prototypes and have around 5 factors to be investigated, which of the following design experiment strategies can be considered the most efficient in finding out the active factors? A) Use a full factorial design, then a Plackett-Burman design using 4 runs. B) Use a Plackett-Burman design using 8 runs, and then a full factorial design. C) Use a Plackett-Burman design using 8 runs, and then a fractional factorial design. D) Use a fractional factorial design, then a Plackett-Burman design using 8 runs. In the paper helicopter three-factor full factorial design experiment, it is found out that 'paper density' and 'blade length' have an active interaction effect. However, both do not have an active interaction effect with 'body width'. Which of the following statements is correct? A) The effect of 'blade length' on the dependent variable (i.e. flying time) is dependent on the level of 'body width', and vice versa. B) The effect of 'paper density' on the dependent variable (i.e. flying time) is dependent on the level of 'blade length', and vice versa. C) The level of 'paper density' is dependent on the level of 'blade length', and vice versa. D) The level of 'blade length' is dependent on the level of 'body width', and vice versa. In the context of Design of Experiment (DoE), what is the maximum number of factors that can be studied using two-level fractional factorial design with 16 runs? A) 17 B) 16 C) 15 D) 4 Select one of the following statements about Machine Learning (ML) which is FALSE: A) Since not all projects are appropriate for ML, one needs to design a project carefully in a way that is suitable for ML. B) ML is not necessarily about saving money, rather about solving repeated problems at scale. C) Data are very important for a successful ML project. In other words, with the wrong data, your ML project can or will fail. D) ML is a type of deep learning that has recently received a lot of attention from the press. Which of the following statements is TRUE about the experimental design for discrete choice data collection? A) The higher the number of attributes varying within a choice-set, the better the design is. B) The attributes should not include more than two levels C) The design should consider an optimal number of choice-sets per survey (or block) to ensure that the respondent does not feel fatigued to answer D) Having prior information about the attributes does not affect the design Which of the following statements is NOT a limitation of revealed preference (RP) data? A) The RP data are not suitable for new products B) The RP data often do not have information on all the choices the individual had while making the decision C) The RP data comprise of actual choices from individuals D) The RP data often consist of insufficient variability across choices to estimate a discrete choice model The Random Utility Theory (RUT) postulates that: A) Individuals' decisions are not affected by their personal (profile) attributes B) Individuals choose an alternative that gives them the maximum utility C) Individuals choose an alternative that gives them the minimum utility D) The utility of an alternative has an unobserved random component only if all attributes are not considered. Which of the following statements is TRUE about the utility function of a Discrete Choice Model (DCM)? A) Utility of an alternative should vary between 0 and 1 B) The higher the coefficient value of an attribute, the lower is the utility associated with it C) Multiplying the utility function with a non-zero value does not affect the probability of choosing an alternative D) Adding a constant to the utility function does not affect the probability of choosing an alternative The Willingness to Pay (WTP) of an attribute for an alternative is higher if: A) the utility decreases with unit increase in the attribute B) the utility is independent of the price of the alternative C) none of the other alternatives D) it requires higher increase in the price for unit increase in the attribute so that the utility remains constant Which of the statements is TRUE about the Discrete Choice Model (DCM) selection? A) Labelled design has better model fit than the unlabelled design B) A conceptually invalid coefficient could be included in the DCM model if it is statistically significant C) The goodness-of-fit of the DCM model does not depend on the loglikelihood value D) All coefficients in the DCM model should be conceptually valid and statistically significant Which of the following statements is FALSE about dendrograms in hierarchical clustering? A) The vertical lines in the dendrogram represent the cut-off distances for different number of clusters B) The vertical axis in the dendrogram is observations C) The horizonal axis in the dendrogram is the distance between observations D) The number of clusters is fixed in the dendrogram Which of the following Quality Function Deployment (QFD) practices best describes how it supports system thinking in an organization? A) It uses analytic hierarchy process which analytically links several hierarchies in an organization. B) It uses the so-called Kano-model mechanism in such a way that the customer needs in an organization are systemically fulfilled. C) It uses a set of competitive assessment matrices for all the internal and external customers in an organization. D) It uses a mechanism through which it is possible for the operators in the production floor to know how their work can be related to the strategic values of the organization. Which of the following statements best describes how the Arrow's impossibility theorem can be useful for Quality Function Deployment practitioners? A) It is best to have a certain target market segment of customers when using QFD. B) It is best to have at least two certain target market segments of customers when using QFD. C) It is impossible to assume a causal relationship, as represented by an arrow, between a demanded quality (the What) and a quality characteristic (the How). D) It is possible to assume a causal relationship, as represented by an arrow, between a demanded quality (the What) and a quality characteristic (the How). In the context of redesigning a product (e.g. a vacuum cleaner), which of the following questions is the most useful for uncovering customer needs according to Ulwick and Bettencourt (2008)? A) Have an improvement-minded attitude, ask "how would you improve the design of this vacuum cleaner?" B) Focus on pain points, ask "what makes using a vacuum cleaner challenging?" C) Start with why, ask "why do you buy a vacuum cleaner?" D) Be straightforward, ask "what do you need from a vacuum cleaner?" Within the context of understanding customer's needs and/or wants, there are four types of pleasure in products (Jordan, 2002), which of the following types of pleasure says that the more human basic senses a certain product attribute can stimulate, the better it becomes? A) ideo-pleasure B) physio-pleasure C) psycho-pleasure D) socio-pleasure Within the context of design thinking, if one aims to improve quality by redesigning a certain service experience where there is little knowledge of customer needs, which of the following research methods is the best to start with? A) Abductive method, e.g. follow up on a critical incident such as a customer complaint. B) Deductive method, e.g. run a customer satisfaction survey to know how satisfied they are with the current service. C) Machine learning method, e.g. use as many similar service experiences as possible and learn from them (no need to reinvent the wheel). D) Inductive method, e.g. run some interviews and observations to understand how they experience the service. Within the context of design thinking as suggested by IDEO, which of the following statements best describes how one may get a good idea during the concepting phase? A) Go for quantity, one must first have many ideas regardless of their quality. B) Ask the heavy users, one must collect ideas from extreme users through in-depth interviews. C) Start simple, one can start with a few ideas as long as they are implementable. D) Garbage-in garbage-out, one must first eliminate bad ideas in the first place. According to the theory of customer dominant logic (CDL), which of the following statements is the most appropriate concerning the value of a product? A) The customer logically will value a product if it has a decent price for the functionality it offers. In short, it is a function of price and its functionality. B) The value of a product is function of brand's image, market values, and customer service. C) The value of a product is a function of, among others, the process of using it, the sensation of possessing it, and its unrealized potential. D) The logic for understanding the value of a product lies in the product's functionality, e.g. a pen is simply for writing. In other words, the value is a function of its functionality. Which of the following statements best describes what happens if one uses performance based questions as follows? Functional: "How would you feel if you were provided with a voice-operated multimedia system that works very well in your hotel room?" Dysfunctional: "How would you feel if you were provided with a voice-operated multimedia system that works very poorly in your hotel room?" Select one alternative: A) It is likely that this new feature will be classified as an attractive attribute. B) It is likely that this new feature will be classified as a reverse attribute. C) It is likely that this new feature will be classified as a one-dimensional attribute although it may be an attractive one. D) It likely that this new feature will be classified as a must-be attribute although it may be an attractive one. Which of the following statements best describes what it means to assume a neutral reference distribution when performing exploratory data analysis? A) To establish a baseline to which the theoretical distribution will be compared. B) To assume a neutral Gaussian distribution as a reference. C) To establish a baseline to which the observational data will be compared. D) To assume a non-Gaussian distribution as a reference. If you have a budget for 16 prototypes and have around 5 factors, which of the following design experiment strategies can be considered the most efficient in finding out the active factors? A) Run a Plackett-Burman design using 12 runs, and then follow-up experiments depending on the result, e.g. a fractional factorial design 2^(3-1) or a full factorial design 2^(2). B) Run a Plackett-Burman design using 8 runs, and then follow-up experiments depending on the result, e.g. another Plackett-Burman design with 4 or 8 runs. C) Run a Plackett-Burman design using 8 runs, and then follow-up experiments depending on the result, e.g. a full factorial design 2^(3) or a full factorial design with replication 2^(2)x2. D) Run a Plackett-Burman design using 8 runs, and then follow-up experiments depending on the result, e.g. a fractional factorial design 2^(4-1) or a fractional factorial design with replication 2^(3-1)x2. In the paper helicopter three-factor full factorial design experiment, it is found out that 'paper density' and 'blade length' have an active interaction effect. However, both do not have an active interaction effect with 'body width'. In other words, the interaction of 'paper density' and 'body width' is almost zero (insignificant), so is the interaction between 'blade length' and 'body width'. Which of the following statements is correct? A) The level of 'blade length' is dependent on the level of 'body width'. B) The level of 'blade length' is independent of the level of 'paper density', and vice versa. C) The level of 'paper density' is dependent on the level of 'blade length'. D) The effect of 'paper density' on the dependent variable (i.e. flying time) is dependent on the level of 'body width', and vice versa. In the context of Design of Experiment (DoE), what is the minimum number of runs needed if one is interested in the main effects of 4 factors and all of the interaction effects? A) 8 runs B) 16 runs C) 12 runs D) 20 runs NRT bank would like to use machine learning to predict the amount of loan (i.e. a number that can go from 0 SEK to 3 million SEK) that they should grant to any given customer. For that purpose, the Analytics department will use data that contain previous customers loan application information as well as the amount that they were loaned. What type of machine learning problem is this? A) Regression B) Classification C) A hybrid type (e.g. a combination of classification and clustering) D) Unsupervised learning (clustering) The parameters included in a Discrete Choice Model (DCM)... A) should have a mean equal to zero B) should have a conceptually valid sign (positive or negative) C) should have a negative sign D) should have a positive sign Which of the following statements is FALSE about a Discrete Choice Model (DCM)? A) DCM is based on random utility theory B) The dependent variable of a DCM is binary, equal to one if a particular alternative is chosen else equal to zero C) All independent variables of a DCM have to be continuous D) DCM is calibrated using maximum likelihood estimation techniques In the context of DCM, the willingness-to-pay (WTP) for an attribute is defined as: A) change in the price such that the other attribute (non-price) can have another level in the choice-set B) change in the price such that the other attribute (non-price) can be completely discarded from the choice-making C) change in the price required to increase the utility by one unit D) change in the price required such that the utility remains the same when the other attribute (non-price) changes by one unit Which of the statements explains the difference between labelled and unlabelled design? A) The choice-set is specific (pre-defined) in labelled design whereas unlabelled design is flexible with any choice-set B) The choice-set is specific (pre-defined) in unlabelled design whereas labelled design is flexible with any choice-set C) Stated preference survey provides an unlabelled design whereas the revealed preference survey provides a labelled design D) Labelled design has more alternatives per choice-set compared to unlabelled design Which of the following is NOT part of a Discrete Choice Model (DCM) design? A) Number of alternatives (per choice-set) and number of choice-sets per block (survey) B) The functional form of the utility function C) Relevant attributes and levels D) Prior information about attributes and levels Which of the following statements is FALSE about the alternatives of a choice-set in a Discrete Choice Model (DCM) design? A) Keep the number attributes to vary across alternatives to a minimum so that the respondent could make a proper choice B) Alternatives should capture the trade-offs among various attributes C) None of the other alternatives D) A single alternative should be dominant in a choice-set Which of the following statements is TRUE about hierarchical clustering? A) Starts with a cluster for each data point, i.e. the number of clusters equals to the sample size (n), then groups the points closest to each other into clusters B) Starts with one cluster for all data points, then divides them into clusters based on the Euclidian distances C) The number of clusters needs to be pre-defined before executing the clustering D) The distances between the clusters should be equal The Arrow's impossibility theorem helps Quality Function Deployment (QFD) practitioners to realize that A) it is impossible to discover a single causal arrow between two quality characteristics in the roof of the house of quality. B) it is impossible to discover a single causal arrow from a design attribute to a customer need. C) an attempt to satisfy all groups of customers may end up with satisfying some of them. D) an attempt to satisfy all groups of customers may end up with satisfying none of them.