operation management

Project Statement

Bass Diffusion Model: New Product Sales and Product Rollover - Q Phone

MGT 207 - Operations Management

Summary

The Q-Phone is an innovative leap forward in the cell phone industry, relying on quantum in place of binary computing. Initial sales of the phone were promising with 263,000 units sold the first year and over 615,000 total units sold two years later. However, sales have been declining for the past ten months. Over the next four weeks, our team will demonstrate that the decline in month to month rate of sales is due to the natural life cycle of the product. We will prove this by building an excel simulation using the Bass Diffusion Model. We will also gather the necessary data to forecast when the product will saturate the market in order to properly time the launch of the next generation of Q-Phone.

Objectives

1. To predict if and when the consumer will buy the product and assign a probability to those events.

2. To assign a timeline to each stage of the Q-Phone’s product life cycle.

3. To find the optimal time to launch the next generation of Q-Phone.

4. To project how market size, innovation parameter, and imitation parameter will impact the sales curve, peak time, and duration of the final product stage of the Q-Phone.

Methodology

Input/output variables:

Months after Q-Phone launch (t)

Q-Phone single purchase adopters (a)

Q-Phone single purchase prospects (P)

In month t, the market potential with respect to number of prospects (Mt)

In month t, the number of sales (Xt)

Net total accumulative sales (Nt)

Single purchase adopter fraction in month t (Ft)

Probability of prospect making a single purchase the following month (ft)

Rate of adoption through innovation (p)

Rate of adoption through imitation (q)

Assumptions:

There is one Q-Phone adopter at time 0.

Adopters purchase one Q-Phone only.

Adopters are influenced by either innovation or imitation only.

Time	Action
Week 1	Estimate the market potential through simulation of one life cycle (60 months) and assign probability to prospects using the random binomial variable function in excel and the Bass diffusion Model: Ft / 1 – Ft = p + q · Ft
Week 2	First, we find the time to peak is tp = t IPt. Then we can use MATCH and INDEX: tp = INDEX(months, MATCH(MAX(Xt), Xt, 0)). Then we find the decline time td =MATCH(0.8,Ft). As the peak time is tp, the starting time is td of the decline stage, and the duration is d of the mature stage. After we simulate one life cycle, we use Data Table to replicate three outputs for N times.
Week 3	To have a sustainable growth, the firm must roll out the next generation, Q-Phone 2.0, before the decline stage. What is the latest time that Q-Phone 2.0 must be launched? Simulate N = 1000 life cycles (sample size).
Week 4	Describe how the changes affect the Sales curve Xt , the peak time tp, and the duration d of the mature stage.

Outcomes

1. The simulation of one life cycle of Q-phone for 60 months will allow us to plot market potential (Mt) and sales (Xt) and accumulative sales (Nt).

2. Find when the maturity stage of the product life-cycle starts (sales pass peak) and how long will it last, simulating 1000 life-cycles.

3. Find the optimal time to launch Q-Phone 2.0

4. Identify possible outcomes given shifts in variables.

Project Guidelines

Each team consisting of four or fewer individuals is required to carry out a project in operations management. The purpose of the project is to deepen and broaden your knowledge and abilities in modeling and simulation. The team will apply the operations methods to a realistic simulation problem. The project topic will be selected by the team and approved by the instructor. The evaluation of the team project will be based on the project proposal, the interim report, the final report and the team presentation.

Project Schedule:

WEEK 6: Initial project proposal (a 2-page project statement) due.

WEEK 8: Interim report (5-7 pages description of the progress of the project, alone with the Excel file) due. Your progress report should discuss in detail the specific problem on which the team has been working. It should contain the detailed description of the problem, the model, the result obtained so far, and the scope of the remaining work.

WEEK 9-10: Team presentations & final report: Submit your final report along with the supporting files (in Appendix). The final report should a) introduce the background and why the investigation of the problem is necessary; b) describe the model and your objective; c) show details of the model construction; d) show the simulation results and include useful statistical tables and graphs; e) discuss your finding, and, f) discuss your conclusion and recommendations. The length of the final report excluding Appendix should be about 10-15 double-spaced pages.

Grading Criteria:

The project will be graded based on the overall quality, scope and difficulty level of the problem. The project will be worth 20% of the final course grade. The following criteria will be used to grade the project:

· Proposal and Interim report: 20%

· Team presentation: 30%

· Final report: 50%

· Model construction: 40%;

· Simulation and statistical analysis: 25%;

· Conclusions and recommendation: 15%;

· Organization/Written of the report: 20%.

Project Proposal Pitfalls

Here is a list of common problems for the project proposal from last year. Your team should try your best to avoid these pitfalls before the submission.

1. The objectives/research questions/problems are not novel/sophisticated enough to qualify for a team project, which accounts for 20 percent of your final grade. The bottom line is, you want to address questions that are interesting and matter to others/management.

2. In practice, analytics is often used to evaluate different system designs (fixed/deterministic inputs, such as different level of warehouse capacity and its impact on the profits), alternative control polices (such as order quantity in the newsvendor model), and sensitivity analysis relative to random inputs (such as the effects of different demand distribution (mean, or variability) on the outputs, the impact of disruption of supply on the supply chain, the effects of uncertain lead time on a transportation system/schedule). A better kind of questions would be what is the optimal design (capacity level, mix of portfolio, mix of transportation modes) or control policy (order quantity, pricing strategy). A pure performance evaluation of a single set of parameters is not meaningful in practice and therefore can hardly qualify for a team project.

3. The models are often not well defined. For example, necessary underlying assumptions are not well spelled out and justified. Input/output parameters are not defined, or their values are not specified. In particular, the input random variables as well as their distributions, their interaction with other inputs outputs are ambiguous at best. As a consequence, the research questions cannot be posed on a solid basis with specific measurements to substantiate it. The bottom line is, if a proposal is not specific and detailed enough for a spreadsheet implementation, it is unlikely to be approved either.

4. The project is not well planned. As a team project, without a clear time table/milestone layout it is hard to deliver a decent project by the end of the quarter.