Decision of Uncertainty: Riordan Manufacturing and Its Web Sites

NB: We do not resell your papers. Upon ordering, we do an original paper exclusively for you.

NB: All your data is kept safe from the public.

Table of Contents

Introduction
Process Description
Problem and Uncertainty
Research
Justification of Research Methods and Data Analysis: Baye’s Theorem
Decisions and Recommendations
References

Introduction

Riordan Manufacturing, a plastics producer with great developmental potential and ambitions, has launched the management restructuring initiative and its IT department has supported it by designing four Web sites. The major issue of uncertainty for the company is thus the probability of each of the four sites to become the most popular among the customers. This issue can be solved using one of the best statistical methods that allow retrieving objective findings from subjective data, i. e. Baye’s Theorem. The calculations this theorem implies reveal that Web sites 1 and 3 have the highest probability for becoming popular, while site 2 has the lowest probability of the kind. These findings allow recommending that Riordan Manufacturing should pay more attention to the development of the sites with the lowest probability of success.

Process Description

Riordan Manufacturing, one of the leaders of the global plastics production industry, has recently developed the strategic idea of a management restructuring. One of the pillars of this idea is the so-called Web-based knowledge management system, for which the IT department of the company has designed four prototypes of Web sites. Every plant of Riordan Manufacturing was provided the access to all the four Web sites, and now the company must identify the most effective Web site design, i. e. the site attended by visitors most often (Riordan Manufacturing, 2010).

Problem and Uncertainty

Thus, the major problem that the current assessment is designed to solve is the uncertainty of Riordan Manufacturing regarding two major factors. First of all, the company needs more specific information about the success of its management restructuring initiative.

Second, Riordan Manufacturing requires specific data regarding the performance of its IT department. Both these factors can be partly examined through the analysis of the attendance of the new Web sites that the IT department of the company has launched in pursuit of the above-mentioned management restructuring initiative.

The four Web site prototypes are designed to serve various aspects of Riordan Manufacturing’s performance including sales, production process, advertising, and customer relations. Therefore, the uncertainty that Riordan Manufacturing faces is all about the aspects of its work that require improvements. Accordingly, the identification of the most popular Web site will allow finding out the most properly developed aspect of the company’s performance, as well as the three problematic aspects of its work (Gilboa, 2009, p. 76).

Research

To operate with the specific data while making any recommendations to Riordan Manufacturing, it is first of all necessary to research the attendance statistics for all the four Web sites in question. For the better convenience of this process, the sites are labeled as follows:

Site 1 (sales) = S1;
Site 2 (production) = S2;
Site 3 (advertising) = S3;
Site 4 (customer relations) = S4.

The research proves that the probability of proper attendance of P(S1 and S3) =.93, while P(S2) =.56, and P(S4) =.75. At the same time, the probability that the largest number of “hits” per minute does mark the popularity of the site (if all “hits” are done by one or several IP addresses or an error is observed in the system) is referred to as M =.15.

Justification of Research Methods and Data Analysis: Baye’s Theorem

Thus, after the research on the uncertainty issue is done, it is now necessary to select the proper fitting statistical analysis model to be used in the given context. Baye’s Theorem can be considered as such a fitting model due to two major reasons. First, Baye’s Theorem allows making conclusions based on subjective opinions, and that is why this theorem is called the model of “subjective probability” (Anderson, 2007, p. 144). Second, Baye’s Theorem enables its users to derive objective probability data from subjective findings of the preliminary research. Accordingly, the use of Baye’s Theorem is the safest way to achieve objectively the inherently subjective results.

Thus, after the basic theoretical framework for the assessment is selected, it is necessary to summarize the currently obtained data and put them in the formula of Baye’s Theorem. So, the currently known information items include:

The IT department of Riordan Manufacturing has launched for prototypes of Web sites (sales = S1, production = S2, advertising = S3, customer relations = S4);
The probability of proper attendance of P(S1 and S3) =.93, while P(S2) =.56, and P(S4) =.75.;
The probability that “hits” per minute do not mark the popularity of the sites is P(M/S1’, S2’, S3’, S4’)= 15% or.15;
The probability that “hits” per minute do mark the popularity of the sites is P(M/S1, S2, S3, S4) = 85% or.85;

Based on these data, the posterior probability according to Baye’s Theorem can be calculated as follows:

P(S1 and S3/M) = P(S1 and S3) x P(M/S1, S2, S3, S4) / P(S1 and S3) x P(M/S1, S2, S3, S4) + P(M/S1’, S2’, S3’, S4’);

P(S1 and S3/M) = (.93 x.85) / (.93 x.85) +.15 =.7905 /.9405 =.84;
P(S2/M) = (.56 x.85) / (.56 x.85) +.15 =.4760 /.6260 =.76;
P(S4) = (.75 x.85) / (.75 x.85) +.15 =.6375 /.7875 =.81.

So, the probability for S1 and S3 to be the most popular is.84, which is smaller than the estimated figure of.93, while S2 and S4 have more chances for popularity than projected -.76 instead of expected.56 and.81 for.75 respectively. Thus, placing all the above-calculated meanings into a joint table will provide better visualization of the data retrieved during the research:

Table 1. Baye’s Theorem.

Fact	Prior Probability P(S1,2,3,4)	Conditional Probability P(M\| S1,2,3,4)	Joint Probability P(A and B)	Posterior Probability P(A\|B)
S1 and S3 are popular	.93	.85	.7905	.7905 /.9405 =.84
S2 is popular	.56	.85	.4760	.4760 /.6260 =.76
S4 is popula	.75	.85	.6375	.6375 /.7875 =.81

Thus, although the actual probabilities are lower than preliminary ones, S1 and S3 still have the highest probability of being the most popular among users.

Decisions and Recommendations

Thus, the above calculations and findings of the assessment allow deciding that the Web sites marked as S1 and S3 have the highest probability for being the most popular among Riordan Manufacturing. Accordingly, it is recommended that Riordan Manufacturing should pay more attention to the development and improvement of S2 and S4 as the Web site prototypes that have the lower probabilities of being popular among the customers. The Web site S2 displays the lowest probability according to the data retrieved through the use of Baye’s Theorem and the customization of this Web site is strongly recommended.

References

Anderson, D. (2007). Statistics for business and economics. Cengage Learning EMEA.

Gilboa, I. (2009). Theory of Decision under Uncertainty. Cambridge University Press.

Riordan Manufacturing. (2010). Official Corporate Web Site. Web.

Click Here To Order Now!