By Jerry Fulkerson
I have one cardinal rule about making decisions…“Never make a big one on an empty stomach.” If your stomach is growling, it’s hard to concentrate on all of the important details. Well, I suppose that isn’t a hard and fast rule, but is does illustrate that most manufacturing engineers have a standardized procedure for making a choice when they are faced with several good machine options for a new project.
In the last blog post, we were identifying the process for our project. In this one we’ll identify the actual machine that we will purchase. The vast majority of machine tools available today are well designed, high performing, pieces of equipment. Nevertheless, making sure that we pick the right machine is still an important part of the equation that leads to a successful project.
Because of that, it’s important to cast a wide net when comparing various pieces of equipment. Just because we based the SWOT analysis on a machine from XYZ Company doesn’t mean that they’re the only, or best, game in town. Unless the need is for very specialized piece of equipment, it would unusual for there to be only one or two machine options.
Going back to the last post, let’s say that our SWOT analysis lead us to using a twin spindle gantry loaded CNC chucker to machine the ACME Widget. Likely as not, our SWOT was based on either a Murata or Fuji machine because they are the market leaders in gantry loaded CNC chuckers. However, for the purpose of making sure that we get the best solution for our part, we also need to consider other strong contenders in that market, including Okuma, DMG/Mori and Takasawa. We’ll ask all those companies to provide a detailed quote proposal for the Widget.
When we bring either the distributor or OEM sales engineers in to ask for a detailed machine quote, we should make sure that each knows, within proprietary limits, everything we know about the project. The last thing that a project engineer wants is a miscommunication that leads to an inaccurate quote. The sales engineer also needs to know whether our company has any special specification needs such as custom color, special pneumatic, hydraulic or electrical requirements, or enhanced safety specifications. This is also a good point to ask for any needed pre-order test runs of the part on the quoted machine and to provide the details of our turnkey run-off requirements .
The information to request that the salesman provide with a detailed proposal can vary based on the project, but at a minimum includes:
- A line item quote that includes all accessories and options required to produce the part(s).
- A brochure of the machine with all specifications.
- A detailed cycle time study.
- A parts manual for the machine (more on this later)
- A diagram of the footprint of the machine, preferably in a CAD format that can be dropped into an existing shop floor plan.
- The results of any test cuts or capability runs you have requested.
- A list of any concerns or conditions that the machine tool builder has with regards to producing the part.
- A list of customer references. I like to get at least five.
Side note 1 – Specifying non-standard components on a machine is costly. Unless they are mandated by a company-wide specification, I seldom go that way. Companies and engineers who do are typically trying to standardize their spare parts inventory or provide commonality between manufacturing locations. That’s useful for a large company with hundreds of machines in its stable, but the savings for a small or mid-size company are marginal and making those requests always increases costs and machine lead time. However, if a change to the standard machine can yield safety, quality or efficiency improvements, it may be worth the extra cost and delivery delay. Items like that should be considered on a one by one basis. For example, while using a substitute brand of directional valve doesn’t improve a machine’s usefulness, adding a safety interlock on a guard may make operation safer. This is a basic cost/benefit decision just like those we make every day.
Side Note 2 – This discussion assumes that we are dealing with a turn-key purchase that includes a machine run-off to show capability to meet quality and productivity requirements. Most companies cede complete control in a turn-key project to the machine tool builder or distributor to avoid the potential of finger pointing down the road. That’s understandable, but I prefer to maintain control of some aspects of the project for either financial reasons or to maintain commonality with existing tooling, program practice, or work holding. Typically, even in a turnkey application I specify and provide:
Spindle and accessory work holding.
Most machine tool vendors will balk at customer supplied materials because they like to maintain full control of the project, charge extra for engineering services, and add a markup to purchased tooling and work holding. Those are all easily understandable reasons, but if your team has the capability and expertise to do some of the work and provide some of the needed components then there is no reason that you shouldn’t stay close to the design process and save costs…provided that you are willing to take responsibility for whatever you supply.
While the cost savings aren’t huge, they can be significant. For example, spindle work holding purchased directly from the chuck manufacturer for a twin spindle CNC chucker might cost $15,000. The identical work holding supplied by the machine tool builder could add $25,000 to the price of the machine. That’s a $10,000 savings.
In a few weeks, each sales engineer will return with his company’s proposal. That’s when we’ll:
- Review any concerns or qualifications that the machine builder has about producing the part to drawing or statistical requirements. If there are issues, now is the time to resolve them. If they can’t be resolved then there is no need to go any further.
- Review the measurement data from test runs, noting any results that are out of specification or below required statistical levels.
- Review the time study and discuss how it might be improved. I try not to make suggestions about how to tool the machine prior to getting this initial time study. Waiting until then gives the application engineer a blank piece of paper to work with. While what he comes up with can sometimes be improved, starting without any predispositions can often produce interesting or innovative options.
- Drop the footprint of the machine into the floor plan to see if it fits well. If not we know upfront that we have some rearranging to do before the new equipment arrives.
- Review the quote on a line by line basis, making sure that it includes any special requests that we’ve made and that we understand the need for all the options and accessories listed. Besides the obvious, this provides an opportunity us to make sure that unneeded extraneous costs haven’t crept in and it allows us to do a feature comparison with quotes from other vendors. For example, if one vendor quotes an internal high pressure coolant system that produces 200 psi and another quotes an external system that provides 1200 psi, a red flag should go up which cues us to ask one vendor why they think 200 psi is enough and the other vendor why they think 1200 psi is needed.
- Test the ability of the vendor to supply spare parts from inventory by using the parts manual that the sales engineer has so graciously provided. Years ago, after having issues with a vendor who seldom had the parts I needed in inventory, I started testing that ability before buying the machine. With parts book in hand and the sales engineer sitting at the table, I call the parts department and ask for price and delivery on ten random items from the parts book. I try to ask for items that I could reasonably expect to need such as, directional valves, pneumatic cylinders, special proximity sensors, belts, filters, seals, and bearings. But, I also throw in a couple of costly items like a spindle or a motor. After the call, I calculate the “fill” percentage. This becomes one of the factors that are considered when making the machine choice.
At this point, we’ve asked the sales engineer a bevy of additional questions and probably identified a number of points that needed clarified. He’s going to go back to the office to take care of all that. This is also the point when many project engineers ask the sales engineer to sharpen his pencil and give them his best price. This is the wrong time and the wrong wording for that and we’ll discuss that in the next post. In a few days you’ll have a semi-final quote. That’s when we’ll start the process of deciding which machine best fits our needs.
When all the semi-final quotes are in, visits to put hands on a machine have been made and customer references have been checked, it’s time to begin the decision making process in earnest. With all the documentation we have and investigation we’ve done, we have an avalanche of information to digest and are essentially in the same situation that we were with the process decision. We need a tool to help us organize and quantify all the data. This time, instead of SWOT analysis, we’re going to use a tool called a (P)roduct (A)ttribute (R)atio worksheet. A blank one looks like this:
As you can see, there are some similarities between a PAR worksheet and a SWOT worksheet. Both include a list of characteristics and give them a weighted rating. The difference comes in what types of characteristics are being considered and how they are weighted. Also note that, with a PAR worksheet, competing machines are listed side by side for easy comparison. This example shows two machines, but ten machines could just as easily be considered by adding extra columns.
Instead of comparing broad advantages and disadvantages of a process, a PAR worksheet drills down and compares detailed specifications, features, or benefits of a particular piece of equipment. It lists the characteristic, the preferred or required specification, assigns an importance (weight ratio), lists the actual specification that the machine or vendor provides, rates that actual value using the scoring key at the bottom of the sheet and then calculates a point value by multiplying the score by the weight. Those point values are then summed for each machine for comparison.
There are three types of specifications considered and each is scored differently. A defined specification is one where an actual value is identified as being a defined limit, such as how much power is need at the spindle. This is scored in four steps from 0-10 as indicated in the key. A comparative specification is one where there isn’t necessarily a preferred value, but there is a way to decide between two or more choices which machine is best, such as measured time in cut. This is scored in three steps from 3-10 as indicated in the key. A binary specification is also a defined limit, but this is one that can be a deciding factor. If a machine doesn’t meet the specification then it’s possibly, but not always, disqualified from further consideration. This is scored as either 5 points or 0 points. The spreadsheet uses conditional formatting to highlight any specification that has a score of zero.
This is an example of a completed PAR worksheet. In practice, there are usually many more characteristics considered than I’ve included here. I’ve had sheets that had nearly one hundred features in the list. Based on the total scores on this PAR, it looks like Machine YYY is the best option to produce our Widget.
Just as it was with the SWOT analysis, completing a PAR worksheet should be a cross functional activity with representation from Engineering, Quality, Maintenance and Safety. Each of those groups is going to have a laundry list of characteristics that their function considers important for the machine.
PAR worksheets are an important and useful tool for a variety of equipment decisions. They not only allow us to organize our data but also provide a way to look at that data objectively. That helps us make a decision that best fits the needs of the project and keeps us on track to “do the job right.” However, my caveat for SWOT analysis is also true for PAR worksheets…gut feelings come into play and a few points in the final score doesn’t compel an engineer to choose one machine over the other. Machine decisions are too complex for that.
Now if you’ll excuse me, I have to decide what movie to watch tonight and I have something important to do first.
If you’d like a copy of the blank PAR worksheet, send me an email here.
Next post – We’ve picked a machine, now it’s time to negotiate the deal.