CUSTOMER NEEDS ELICITATION FOR PRODUCT CUSTOMIZATION

1
CUSTOMER NEEDS ELICITATION FOR PRODUCT
CUSTOMIZATION

• Yue Wang
• Advisor Prof. Tseng
• Advanced Manufacturing Institute
• Hong Kong University of Science and Technology

2
Background
Axiomatic design
CNs are expressed in explicit product
specifications.
3
Introduction

• Customer needs elicitation should be
• Good predictive, customer insight
• Fast for customers and for designers
• Cheap reduce market research cost
• Easy reduce drudgery and errors

4
Research issues

• Can we find what people want quickly and
  inexpensively?
• How to avoid confusing customers with too many
  products?

5
Challenges

• Customers are
• Impatient to specify a long list of items
• Unable to articulate their needs
• Unaware of latent needs
• Lack of information about available options
• Interlocking among attributes

6
Approach

• Research framework
• Bayesian network based preferences representation
• Adaptive specification definition procedure
• Recommendation for customized product

7
Preferences Representation

• Uncertainty of the purchasing choices
• Customers are heterogeneous
• Choice decisions differ under various situations
• The context of purchase differs
• Dependency among preferences towards different
  attributes

8
Preferences Representation

• Bayesian network

9
Specification Definition

• The important considerations in this phase
• Customers are not patient enough to specify a
  long list of items.
• The items differ a lot in terms of the amount of
  information they can provide.

10
Specification Definition

• Basic ideas
• Present the most informative query item to
  customers
• The value of information
• the additional information
  received about X from getting the value of Yy.

11
Specification Definition

• The solution for f (Blachman, 1968)

N. M. Blachman, The amount of information that
y gives about X, IEEE Trans. Inform. Theory,
vol. IT-14, no. 1, pp. 27-31, Jan. 1968
12
Recommendation

• Given
• Customers preferences information
• Determine
• Which products should be recommended?
• In what order to present the recommendations if
  more than one recommendations are presented?

13
Probabilistic relevance computation

• Probability of relevance under binary independent
  assumption
• Probability of relevance considering first order
  conditional dependency

14
Probability ranking principle

• The idea is to rank products by their estimated
  probability of relevance with respect to the
  information obtained.
• Probability ranking principle is optimal, in the
  sense that it minimizes the expected loss.

15
Schematic framework
16
Evaluation metrics

• Precision rate
• Recall rate

17
Evaluation results

• The recommendation based on probability ranking
  can guarantee the highest precision and recall
  rate.
• If customers preferences to all the components
  are independent and the potential preferences
  towards all the alternatives of an attribute are
  random, the specification definition method based
  on the information gain has the highest precision
  and recall rate.

18
Evaluation results
19
Evaluation by utility

• Preliminaries
• Stochastically dominate
• If , then approach 1
  stochastically
• dominates approach 2.

20
Evaluation results

• The presented method
• stochastically dominates other approaches.
• is optimal with respect to any nondecreasing
  utility function.

21
Summary

• An approach to elicit customers preference is
  presented.
• The model can be used to adaptively improve
  definition of product specification for custom
  product design.
• Based on the model, customized query sequence can
  be developed to reduce redundant questions.
• Product recommendation approach is adopted to
  further improve the efficiency of custom product
  design

22

• Thank you!

Your suggestions comments are highly
appreciated!
23
Extension to binary independent assumption

• Theorem A probability distribution of tree
  dependence Pt(x) is an optimal approximation to
  P(x) if and only its maximum spanning tree.
  Chow and Liu, 1968

24
Why customized product design

• Well calibrated customized product design can
  integrate customers into design activities
• Mitigate the side effect of sticky information
• Better meet customers requirements
• Loyalty can be enhanced.
• Help identify latent needs guide future product
  development

25

• Lemma 1 Suppose approach 1 proposes n
  recommendations in a sequence S1(r11,r12,r1n).
  Each recommendation r1i has probability p1i to
  meet the customer needs. The sequence is arranged
  such that
• . Approach 2 also
  proposes n recommendations in a sequence
  S2(r21,r22,r2n). These n recommendations may be
  different from the ones in sequence S1.
  Similarly, we also have corresponding probability
  serial and If
  for all , then X1
  stochastically dominates X2 where Xi is an
  indicator of the number of satisfactory
  recommendations by using approach i.

26

• Lemma 2 Suppose approach 1 proposes n
  recommendations in a sequence S1(r11,r12,r1n).
  Each recommendation r1i has probability p1i to
  meet the customer needs. The sequence is arranged
  such that . Approach 2 also
  proposes n recommendations in a sequence
  S2(r21,r22,r2n) which is a permutation of
  S1(r11,r12,r1n). Then the distribution of
  satisfactory product for approach 1 is identical
  to approach 2.

27

• Lemma 3 Let U(x) be a nondecreasing utility
  function where x is the number of satisfactory
  recommendations. Let Xi be an indicator of the
  number of satisfactory recommendations by using
  approach i. If X1 stochastically dominates X2,
  then the expected utility by adopting approach 1
  is greater or equal to that of approach 2, i.e., .

28
Evaluation

• m the number of attributes
• ni the number of alternatives of the ith
  attribute
• N the total number of configurations
• Pijk if the jth alternative of the ith component
  is selected, the probability that the kth
  configuration is the desired one.
• The entropy of the configuration space if the
  jth alternative of the ith component is selected
• The expected entropy of the configuration space
  if the ith component is proposed for a customer
  to specify

29
Background

• Competitive and changing market
• Shorter product development time
• Product variety proliferation
• Bigger penalty cost of failing to meet customers
  needs or catch up customers needs changes

30
(No Transcript)
31
Probabilistic relevance model

• Probability of relevance (including first order
  conditional dependency)
• Parameters setting

32

• tailor product to different needs
• how to avoid confusing customers with too many
  products
• Can we find what people want quickly and
  inexpensively
• how to find out if a customer is interested in a
  virtual which doesn't exist
• reducing inconsistent preferences
• good predictive, customer insight what people
  buy or how many will people buy it
• fast for them and for us it should be fast,
  doesn't cost so many time
• cheap reduce market research cost should be
  cheat
• easy reduce drudgery and errors should be easy
  for both customers and designers
• That's all the questions in marketing science
  today.