Page 22 - MMI-SeptOct2021 SM upload

object or an environment. They tion and have the deep critical modeled and evaluated, and
then use this data to perform thinking and problem-solving it is even more difficult to rep-
analysis to determine how to skills required to optimize the resent them graphically or
optimize complex products or process or product design. with objects.
systems. This simulation can oc- The result of well-applied Using DES, these high-com-
cur ahead of any physical action simulation can include a plexity systems can be modeled
such as the building of a proto- range of benefits, including and analyzed to help make
type or the establishment of a speeding time to market, creat- business decisions and drive
manufacturing line. ing better, more reliable prod- improvements in various ways.
It is an invaluable tool for ana- ucts and more. In some cases, and applications,
lyzing and optimizing dynam- The three types of simulation DES is the only way to study
ic processes. This is especially that are used for design, man- a system and make improve-
true when mathematical opti- ufacturing, or logistics simula- ments due to the fact that the
mization of complex systems tion are Discrete Event Simula- system is so large and complex
is impossible using traditional tion, Design for Reliability, and that any other method will
methods – and when conduct- Finite Element Analysis. Each take years to model, statistical-
The three ing experiments within real type addresses different opti- ly study and then make need-
types of systems is too expensive, time mization challenges. ed changes (assuming the re-
simulation consuming, or even dangerous. al-world system still exists and
that are Because it involves running an Discrete Event Simulation has stayed the same since the
used for extensive range of scenarios, For any product, understand- study started).
design, simulation provides engineers ing the steps and interactions Since DES is a representation
manufac- with a controlled and precise of multiple processes is key of a physical system, it allows
turing, or methodology in support of ob- to optimizing production. the testing and experimenta-
logistics jective decision-making. When modeled, simulated, tion of an existing or current
simulation The most critical consideration and analyzed, these steps of- system under various scenari-
are Dis- for any simulation project is to fer important insight into the os and provides insights based
crete Event ensure a clear understanding overall process. This practice on those scenarios. This is
Simulation, of the variables, constraints, in the Manufacturing industry a very powerful concept. In
Design for and information needed. This is is called Discrete Event Sim- theory, a real-world model is
Reliability, most often the longest part of a ulation (DES) and is probably being copied in a software-
and Finite simulation initiative. Ultimately, the most frequently used type created environment, creating
Element the quality of the data input plus of simulation. a ‘digital twin.’
Analysis. the engineers’ expertise deter- DES models real-world phe- Using this digital twin, DES
Each type mines the quality of the results. nomena or a system of opera- allows for the possibility of
addresses Simulation analysis yields the tion as a sequence of discrete ‘what-if’ scenarios to be eval-
different best results when performed events. Real-world phenomena uated. Because the system is
The following diagram shows how simulation engineers may choose to model and analyze high-risk areas before
optimiza- by a dedicated team of engi- with stochastic elements can- being represented by a model,
production and recommend solutions to help ensure product reliability targets.
tion chal- neers who specialize in simula- not always be mathematically it is easier to experiment and
lenges.
DfR is a multi-step process that takes many approaches
DfR is a multi-step process that takes many approaches

3. Analyze

1. Identify 2. Design 5. Validate 6. Control

Source: Flex Ltd
4. Verify

At the core of DfR is the physics of failure, which includes testing to identify issues and statistical analysis
22 | September-October 2021 Modern Manufacturing India
to determine reliability prediction. This data can then be used to recalibrate as needed prior to
physical prototyping.

Mean Time Between Failures (MTBF) is an important DfR tool, which predicts elapsed time between inherent
failures of a mechanical or electronic system during normal system operation. MTBF is calculated as the
arithmetic mean (average) time between failures of a system. This analysis is used for repairable systems.
Mean Time to Failure (MTTF) is the expected time to failure for non-repairable products or systems.

Using MTBF calculations, an engineer can improve a design by understanding points of failure.

For example, an engineer wants to evaluate 144 components found in a bill of materials for planned usage in
a printed circuit board. Using calculations under different operating temperatures, the engineer determines
that 20 of the 144 parts are identified as high risk. The engineer then submits these parts for consideration to be
replaced by parts that perform well at the higher operating temperature.
Another example is determining that 53 parts in a design will last ten years, while 11 will last for three years. This
finding indicates a need to replace the 11 parts to be comparable with the longer lasting parts.

DfR provides the simulation needed for:

• Cost control: On average, a considerable portion of a project’s budget is typically allocated
to design

• Preserving profits: Products get to market earlier, preventing erosion of sales and market share

7 | MANUFACTURING REACHES A TIPPING POINT