Relcode

Snapshot

RELCODE, using, for the most part, Weibull mathematics, determines the probabilities of component-failure. It’s for equipment "parts" and helps you decide whether to replace them preventively or only-on-failure...and if preventively, exactly when.

It tackles, and solves, the thorny dilemma that all maintenance managers face...

"If we replace the part now, maybe it would have lasted quite a while longer...but...if we wait for a scheduled overhaul, the component may fail on-the-job— and cause a costly emergency repair".

RELCODE gives you the "odds"— the probabilities— of a component failing at various hours-of-service intervals as well the most economical replacement-interval (3600 hours, for example). The higher costs of making a replacement at other than the optimal interval is also provided. The maintenance manager can then make confident cost-related decisions, to replace or not to replace, based on the program’s output.

The Weibull analysis also give you the failure pattern of your components…w burn-in, w random, w wearout — which indicate the root cause of failure — and the mean time to failure (MTTF), and other reliability parameters.

What Data
Do You Need?

RELCODE needs two inputs from the software-user.

(1) DATA FROM YOUR RECORDS.
One of them is the life-history of each machine-part being analyzed. That is, how long did each bearing— or belt, or breaker, or shaft— last before it was replaced on account of failure or any other reason. This hours-to-replacement data forms the main part of RELCODE’s required input.

But what if you’ve just begun to collect this kind of information? RELCODE arrives at good answers with very limited amounts of input. As time goes on of course, you’ll add to your data-bank and RELCODE will use the additional data to come up with even tighter recommendations for you.

(2) EXPERIENCE-BASED ESTIMATES.
The only dollar-figures the program needs are your estimates of...(a) The cost of carrying out a preventive replacement of the component you’re analyzing, and...(b) The cost of replacing the component in the event of an on-the-job failure. Your maintenance records and your maintenance experience should give you these numbers, and — more good news — they needn’t be of decimal-point accuracy to obtain good results.

The
Output
You’ll Get

One of the many nice things about RELCODE is that it can be used by managers who simply want to obtain basic guidance — i.e., when is the optimal time to replace the part, but it can also be used by engineers and reliability specialists to get a very complete statistical picture of a replacement scenario.

FOR THE LINE-MANAGER, RELCODE shows the number of hours — or revolutions, or cycles (you get to pick the utilization-unit you want) — that will give you the lowest-cost solution to the "when should we replace" question. You’ll also obtain the costs associated with following RELCODE’s recommendation versus the costs of letting the part run to failure.

Suppose though, that you don’t want to incur the often-excessive costs associated with a component-failure, but you also don’t want to carry out the maintenance exactly when RELCODE says you should? You want to do it sooner — or perhaps later, because of other considerations like a general overhaul or a plant shutdown.

Handling that question is easy. You simply tell the program what your preferred maintenance interval is, and RELCODE will come back and show you whether the cost-penalty for not following it’s optimal plan is high or low. If it’s high you’ll want to re-consider the advantages of an earlier or later replacement. If it’s low you’ll probably want to use your other criteria.

RELCODE delivers even more to the line-manager.

For example, the output he or she needs is shown not only numerically, but it’s almost always illustrated graphically as well. Some people like numbers, some like pictures. RELCODE provides both!

Finally, this software package delivers still another bonus — it works out the inventory of spare parts you’ll need to carry in order to meet predicted replacements.

FOR ENGINEERS AND RELIABILITY SPECIALISTS, RELCODE supplies a veritable flood of information. Only a personal look at the program can demonstrate its very wide range of features, but here are just a few of them:

FOR BOTH THE LINE-MANAGER AND THE SPECIALIST,
RELCODE provides the most vital information of all — the root cause of failure — the "nub" of. managing the failure situation.

It does this by identifying the failure pattern.

If you experience an increasing hazard rate that means there’s a normal "wear-out" condition which is usually dealt with by a preventive replacement policy. If there’s a decreasing hazard rate you have "burn-in" where weak or faulty parts fail early, and this means there are problems with the manufacture or assembly of the component. (Get after your supplier!). If there is a constant rate you have a "random" failure condition which indicates that failures are being caused by some external factor such as debris entering a pump or excessive loads on a conveyor.

To simply replace machine components is one thing, but to know why the failures are occurring is another. RELCODE tells you "why".

In summary...whether you’re a line-manager who just wants "the answer" without a whole lot of statistical measures, or a reliability specialist whose job it is to analyze parts performance in-depth, RELCODE will give you want you need. No more — no less.

This graph shows the probability of a component lasting to a given age. For example, th probability of a new component lastin to 3900 hours is 65%.

This display is the mirror-image of the previous graph. It tells us that if the component has run for 3900 hours, there is a 35% chance of it lasting longer.

The Hazard Function shows the failure rate among items which have survived to any given age. The increasing failure rate (as in the example) indicates that this component is subject to wearout failures.

This is one of the most "instantly-useful" outputs of the entire program. From the curve, and from the summary in the box, you can get an immediate picture of the replacement options for this component. You can see that the best time to replace, when the costs are lowest, is in the 1300-3900 hours range — 2080 hours is the very vest solution — and you can see that replacing earlier is likely to result in substantially higher costs. Replacing later, on the other hand, doesn’t increase your costs all that much.

This display summarizes some vital information and does it very neatly indeed. The Preventive and Failure Replacement costs at the top of the screen is your input, and is used in the calculations of the output, lower down. The output show that the optimal replacement age is 2080 hours and it shows the relative costs of preventive replacement versus running to failure — along with the savings you get by using the optimal policy.

RELCODE also allows you to use your own "specified" replacement age — it may just not be convenient to replace at 2080 hours. In this example it was decided to replace at the 2000-hour mark, and the output shows the consequences of doing this. Of course, the more one deviated from the recommended (optimal) replacement age, the greater the costs will be. Yet, in many instances it may be the "correct" decision, given other maintenance scheduling factors.

An "added bonus" with RELCODE is it’s Spare Parts Requirements calculations. It’s an extremely useful guide to help decide how much (costly, and resource-demanding) inventory you must carry. This feature alone is worth the price of the software.