Mobile Edition
Print
Get the Mag
Weekly UpdatesMedia matters
Q: Our client is a major metropolitan newspaper and the goal is to boost profit by improving circulation. The plan is to use some form of the newsvendor problem to predict sales at newspaper stands.
The issue is that the selling price of a copy of this newspaper is much less than the cost to make and distribute the paper. Of course, this company makes up the loss by advertising profit, but they argue that advertising has no correlation to circulation. In order to use the newsvendor model, though, the price has to be greater than the cost for a single product.
I'm sure this problem has been confronted in the past. Can you point us in the right direction?
A: First, you cannot use the newsvendor model to "better predict sales at newspaper stands." The newsvendor problem requires as an input a probabilistic description of the number of papers demanded (presumably by location in your application). Second, you cannot boost profits by selling something at a price that is less than the cost of making and distributing it.
Thus, if there is some monetary value associated with the sale of individual papers (e.g., for each 1,000 papers sold at newsstands, the company anticipates an increase of $X in advertising revenue), then $X/1,000 should be added to the sales price of each paper as this is implicitly additional revenue that accrues to the company as a result of paper sales.
One problem with this may be that the relationship between newsstand sales and advertising revenue may not be well-known. The relationship may not be linear, which will further complicate the use of the newsvendor model in determining how many papers to allocate to each newsstand.
Finally, in the multi-location problem there may be substitution effects in the demand for papers. In other words, if the first newsstand that I pass on my way from the train to the office does not have the paper, I will try to buy it at the second newsstand I pass, and then at the third, and so on.
MARK S. DASKIN, PH.D.
Going lean no matter what
Q: We are working on a lean transformation of a low-volume/high-mix electromechanical production facility. Currently we work in an "artisan" environment. Supplier on-time delivery is not good. Critical parts shortages are common. We think a flow cell would reduce WIP and total time in the system. Is it possible to move forward with lean flow even with continuing supplier issues?
The work-around is to build a lot of WIP as partial builds, waiting for the one last part to add at the last minute. It causes all the expected issues of confusion and not having the right part available for the now available units.
A: The flexibility and shortened lead-time with lean flow can tremendously help your operations. Right now you have to build WIP and hold unfinished pieces so that you can quickly turn them around and get the parts out the door. With lean you can reduce the overall lead-time. With critical part shortages, it is still going to be difficult to manage schedule production unless you can work with your suppliers to improve on-time delivery.
Flow production is a philosophy that rejects batch, lot or mass processing as wasteful. Product should move (flow) from operation to operation in the smallest increment, one piece being ideal. Product should be pulled from the preceding operation only as it is needed. Often referred to as "one-piece flow," only quality parts are allowed to move to the next operation.
Changing from an artisan work environment to a cross-trained work force is also critical to achieving lean. Multi-skilled operators are associates at any level of the organization who are diverse in skills and training. They provide the organization with flexibility and grow in value over time and are essential for achieving maximum efficiencies for JIT.
In a low-volume/high-mix production, a key lean tool will be single-minute exchange of dies (SMED). SMED is a method of increasing the amount of productive time available for a piece of machinery by minimizing the time needed to change from one product to another. This greatly increases the flexibility of the operation and allows it to respond more quickly to changes in demand. It also has the benefit of allowing an organization to reduce greatly the amount of inventory that it must carry because of the improved response time, while maximizing return on investment and economic value-add.
ELIZABETH CUDNEY, PH.D.
Find value in training, internships
Q: I am currently a graduating senior at California Polytechnic University San Luis Obispo majoring in industrial engineering. I'm looking for ways to get some experience that will look good on my resume, as well as provide benefits once I begin work in industry. I'm thinking about trying to get certifications through APICS for production and inventory management and possibly a lean Six Sigma type of certification.
Do you have any recommendations on whether this is worthwhile, and, if so, the most relevant ones to explore?
A: A good way to get experience is to find an internship with a company. Many organizations use internships to evaluate prospective employees. Hopefully the professors at your university will be able to suggest some good organizations to contact. Certificates such as Six Sigma green belt, lean, etc., are also valuable. When there is interest, IIE has provided weekend training classes on these topics at several universities. If your school is interested, please have them contact me. We provide this student-only training at a steep discount over what we charge practicing engineers.
LARRY AFT, P.E.
Learning for lost time
Q: I've got a three-year diploma in industrial engineering from an Irish college. I've worked in various industries. I haven't worked directly in the field for more than 10 years and I'm now interested in getting back into industrial engineering. How can I update my skills along with getting my diploma recognized in the United States?
A: Regarding updating your skills I would recommend continuing education courses. Here at IIE we offer many courses--some of them online--that would let you sharpen your skills. Information about these is available at www.iieseminars.org.
Regarding having your diploma recognized, that is a more difficult question. Each state has its own rules and regulations. The best advice I can give you is to contact the Engineering Registration Board in any state that you might want to live and work in. They would give you their requirements for recognition of the diploma. Here is a link to a listing of these boards: www.ncees.org/licensure/licensing__boards/
I hope this helps.
LARRY AFT, P.E.
Reduce defects with 'nextmarks'
Q: I have been trying to research what the typical scrap rate is for the electronics industry. Do you have any information of this nature or can you direct me to where I can find it?
A: I doubt if there is something like a generic reject rate for the electronics industry. We know Motorola decided to go for 3.4 defects per million as their mark when they pursued their Six Sigma program. There can be one reject rate for the components from suppliers, a different reject rate for the internal processes within the company and a third rate for the final product in the hands of the customer. Motorola's rate was for the final customer.
Rather than benchmark using a reject rate, I suggest doing what is called "nextmark." In nextmark, you aim for the ideal reject rate--which naturally is zero defects--and then using root cause analysis and all other tools of Six Sigma, see how close you can come to no defects at all.
Another practical way to reduce defects is also to begin with what you already know as your baseline defect rate, and then aim to reduce that rate by 50 percent in each rapid improvement event that you undertake.
MERWAN MEHTA, PH.D.
FEED