For eleven years I've been signing off on manufacturing process decisions, and I've made enough expensive mistakes to fill a binder. When I started as a process engineer in 2014, I thought choosing a manufacturing route meant finding the best technology. Now I know it means choosing the least-bad compromise for your material, quantity, and tolerance. That lesson cost me roughly $47,000 before I believed it.
I'm not going to tell you there is one right answer. Instead, here are three scenarios I use when I look at any new part. Each is based on something I got wrong at least once.
Scenario A: Tight-tolerance metal parts with short or medium runs
When the part needs to survive load, heat, or inspection and the material is metal, machining is usually the default. A vertical machining center (VMC) or turning center with good tooling is the workhorse. If you are in this scenario, the ISCAR catalog is not a coffee-table book. It is a technical reference for insert shapes, grades, and cutting data.
In 2021, I got that wrong on a 300-piece order in 17-4 PH stainless steel. In a hurry, I selected an insert designed for softer material because it looked close enough. By part 15, the cutting edge let go. We scrapped 285 parts, lost $3,100 in material, and added almost two weeks to the schedule. The failure was not the metal. It was my tooling choice. I should have matched the insert geometry, grade, and corner radius to the workpiece.
One terminology side note: if you are searching what is a VMC drink, that is not a term you will find in machining standards. In metalworking, VMC stands for Vertical Machining Center. The beverage version belongs to cocktail and social-media conversations, not cutting tool specifications. The larger lesson is the same as my insert mistake: verify the term and the requirement before spending money.
Scenario B: High-volume plastic parts and the fully electric injection molding machine
When annual quantity climbs past the low thousands and the material is thermoplastic, injection molding becomes a serious option. That brings up the question I hear constantly: should the press be hydraulic, servo-hydraulic, or fully electric? I am not going to tell you that hydraulic machines are obsolete. For deep-draw parts, very high tonnage, and existing tooling, they still make sense. What I will say is that the cost comparison has to cover more than the purchase price.
In 2022, I argued against a fully electric injection molding machine because it cost $38,000 more than a hydraulic alternative. The first month of production looked fine. By the fourth month, extra rejects, settling time, energy consumption, and cooling-water load had eaten most of that saving. According to Arburg (arburg.com, 2023), all-electric machines can cut energy consumption by up to 50% compared with hydraulic machines in many applications. They also provide tighter clamping repeatability and faster response. That matters for medical, automotive, and precision consumer parts.
The blind spot I keep seeing is focusing on clamp tonnage and sticker price while forgetting that the mold itself must be designed and machined correctly. A fully electric injection molding machine cannot fix a mold cavity cut with the wrong tool path or wrong runner layout. If you make molds, the same tooling logic applies: insert selection affects every single cycle.
Scenario C: Low volume, complex plastic geometry, and frequent design changes
For one to fifty plastic parts, especially when the design is still moving, 3D printing is the right starting point. The 3D printer FDM meaning is simple: FDM stands for fused deposition modeling. In the ISO/ASTM 52900 taxonomy, it is classified as material extrusion. A filament is heated and deposited layer by layer. It is not the only 3D printing process, but it is the most common.
FDM is strong for fit checks, ergonomic prototypes, and shop-floor fixtures. It is weak where layer adhesion, smooth surfaces, or isotropic strength matter. Most people outside the industry assume printed plastic is printed plastic. It is not. A printed ABS bracket may be fine in an office, but in a warm machine area it can soften and creep.
I made that exact mistake in 2023. I approved a bracket printed in ABS for a fixture near a motor housing. Within a week, the bracket sagged, the sensor it held moved, and the line stopped for four hours. That cost roughly $900 in downtime. FDM was not the villain. I had chosen the right process for a prototype but the wrong material and application for the final function.
How to tell which scenario you are in before you spend money
I now use a three-question pre-check list: What is the material and expected load? What is the annual quantity? What tolerances and surface finish does the drawing require? In that order.
- If the part must be metal and hold tight tolerances, start with CNC machining and choose tooling from the ISCAR catalog by matching the operation, insert geometry, and cutting conditions to the workpiece.
- If the part is thermoplastic and volume is high, put a fully electric injection molding machine on the shortlist, but model the total cost of energy, maintenance, mold-making, and quality risk before making a decision.
- If you need a few parts quickly and design changes are likely, print them on a 3D printer. Know the 3D printer FDM meaning and its limits.
Looking back, the most expensive phrase in my planning vocabulary has been, this should be fine. The fix is simple: write down the most important requirement and check the process choice against it. If the requirement is metal and tolerance, do not print. If the requirement is a million plastic parts, do not machine one by one.
Since the start of 2024, this checklist has caught 47 potential process errors in our team. The dollar amount saved is well above the $47,000 I lost before I learned to use scenarios instead of instinct. Efficiency is not only about faster machines; it is also about not repeating avoidable mistakes.