Oh, oh! The engineers are in a pinch. At the most recent staff meeting one pf the topics discussed was that one of the newer light fixtures was having some problems. This is the model L799AC. They are being returned at a rate that is much higher than anticipated. It appears that the L799AC is just are not as durable as customers are expecting from a light fixture. The planned MTBF was some 1,500 hours. The Operating time until failure for the returned units ranged from 99 to 199 hours. That appears to be an eye-watering difference! Customers and the LUML engineers know that there is a “mixture” of factors (e.g., reliability; maintainability even though is most cases customers plan on doing very little maintenance beyond changing the bulb once the fixture is installed; and availability). The CEO asked them to look into this situation, and to send her a memo on the situation. Alas, the engineers are not always agreeing on what the situation is or how to proceed. With somewhat bruised egos, then have asked your boss Tom Perkins (VP/Director of Logistics) for assistance in analyzing the situation and drafting the memo to reply to the CEO.

 Q#1. (100 points) The idea of a “bathtub” is often used to describe the failure rate curve.  When searching the Internet and other sources (some old textbooks) they got a zillion hits (or at least a very large number). But what they need is a concise answer to explain this 3 concept to the non-engineer, non-logistician CEO. Probably a figure. And then a brief explanation of the elapsed time and the failure rate. A time period/region (or is that periods/regions?). And the characteristics of each period/region that the curve covers. 

 Q#2. (100 points) Luckily several customers kept records on failure of the fixtures. Here is a summary of the 8 units returned due to failure. Returned  Unit # Operating time until failed (hours) 1 133 2 155 3 199 4 176 5 148 6 165 7 160 8 99 [NOTE: This is obviously a small sample of 8 observations. But treat this as if it represented a very large sample so, for example, you do not need to try to apply a finite population correction factor, etc.] Q#2a. If the returned fixtures were in the stage where the exponential reliability function applied, based on these actual MTBF’s what was the theoretical reliability at 200 hours of operation? [Note: to 2 decimal places.] Q#2B. Show your work and explain your answer.

 Q#3. (100 points) The fixture has three major components which act in series. [Note: final answers to 2 decimal places.] Part #1 has a reliability of 0.99. Part #2 has a reliability of 0.97. Part #3 has a reliability of 0.90. Q#3a. What is the theoretical reliability of the system/product? Q#3b. If LUML is going to work on improving the reliability of the system/product, which part do you recommend the engineers initially concentrate on? Why? 4 Q#3c. If LUML can improve the reliability of that part by 5 per cent, what should the new reliability be for the system/product?  

Q#4. (100 points) The planned MTBF was 1,550 hours. The average corrective maintenance time was planned to be 1 hour. [Note: satisfactory conditions, etc., etc.] Q#4a. What was the planned inherent availability of the system/product? Q#4b. Explain your work and the implications for LUML. 

Q#5. (50 points) One aspect of MGT5061 is to address “policy”. Give and explain two policies concerning RMA that LUML should consider implementing. Include the reason(s) why the policy would be appropriate? (Or, if you prefer, consider it RAM knowing RAM distorts the relationship to some degree.) No real limits on length. On the one hand, I am not looking for multiple pages. On the other hand, you do need to cover the topic. Usually a few sentences with solid reasons for each is sufficient. 

Uploading the files attachment with the actual assignment and notes giving  



Source link