6. Manufacturing 3 - Module 06: Hitachi’s DFA Method

Hitachi’s DFA Method

Overview

• Hitachi's Design for Assembly (DFA) method is specifically for manual assembly.
• Lecturers: Dr. Arfauz A Rahman and Prof. Yan Jin.

Learning Outcomes

• Understand the basic principles of Hitachi’s DFA method.
• Learn through examples how to apply Hitachi’s DFA.

Background

• Hitachi’s DFA method was developed in the late 1970s as part of Hitachi’s initiative to create products efficiently assembled by automation.
• First developed in 1976 in Japan.
• Also referred to as the “Assemblability Evaluation Method (AEM)”.
• Based on a quantitative evaluation approach.
• Fully described by Miyakawa and Ohashi (1986).

Assemblability in Hitachi’s DFA

• Interpreted as “assembly producibility”.
• Assemblability evaluation is built around the “assessment of assembly operations”.
• The essential principle is “one part – one motion”.
• For each part of the assembly, there should be only one straightforward and simple motion required to fit and secure it.
• An assembly scoring system is used:
  • A simple straight motion scores with no losses.
  • A more complex motion scores with progressively greater losses.

Operation Element Symbols

• Symbols represent the direction of motion of a part.
• Symbols represent fixture and forming requirements.
• Symbols represent joining and processing requirements.
• Every assembly motion is assigned a “penalty point”.

Performance Indicators

• Assemblability evaluation score (E):
  • Used to assess design quality, reflecting the difficulty of assembly operations.
  • Ranges from 0 to 100.
  • A relative measure of how producible a design will be in production.
    • $E = 0$ (infinitely hard assembly).
    • $E = 30$ (hard to assemble).
    • $E = 80$ (easy to assemble).
    • $E = 100$ (ideal assembly).
  • Higher $E$ is better.
• Assembly cost ratio (K):
  • Used to estimate improvements in assembly costs after design changes.
  • For example, $K = 0.74$ means the new design will cost 74% of the original to assemble.
  • Lower $K$ is better.
• Assembly time (AT):
  • Measured in T-downs.
  • One T-down is the time taken for one downward movement with a part.
  • The T-down value is determined for a particular factory to reflect true assembly costs.
  • If the T-down value is unknown, it can be used as a relative measure.
• Candidate for elimination (CFE):
  • Determines whether a part is a CFE based on:
    • Motion: Does the part's motion cause difficulty in assembly?
    • Material: Does the part need to be a different material from the rest?
    • Service: Does the part require maintenance?
  • If a part is a CFE, a value of “1” is assigned to it for use in the evaluation form.
• Part count design efficiency (PCDE):
  • A measure of the design efficiency of the assembly.
  • Depends on the CFE, the quantity of parts ($n$), and the sum of the quantity of parts ($N$ or $\Sigma n$).
  • Target to achieve a value of “1” in redesign, indicating no CFEs.
• Simplicity factor (SF):
  • An overall efficiency measure of assembly.
  • Depends on the assemblability evaluation score ($E$) and PCDE.
  • If $E$ scores of the original and redesign are similar, use $SF$ to judge assembly efficiency.

Relevant Equations

• Assemblability evaluation score: $E = N \times 100 / AT$
• Assembly cost ratio: $K = AT / AT_{ORIGINAL}$
• Assembly time: $AT = \frac{\Sigma [T(n)]}{100}$
• Part count design efficiency: $PCDE = M / N$
  • Where $M = N - \Sigma [CFE \times n]$
  • $N$ is the sum of the quantity of parts, $\Sigma n$
  • $AT$ is assembly time

Calculating $T(n)$

• $T(n) = T \times n$
• Where $T = \alpha (100 + \Sigma e)$
  • $\Sigma e$ is the total penalty due to assembly (refer to the penalty points of operation elements).
  • $\alpha$ is a corresponding factor value dependent on $m$ (number of operational elements).

• The "Alpha" equation: $0.85 + 0.15m$ is only applicable in redesign.

Procedure

1. Preparations:
   • Prepare products to be evaluated, such as conceptual drawings, design drawings, assembly drawings, and samples.
   • Prepare the assemblability evaluation form (referred to as the “evaluation form”).
2. Attaching Operation Analysis:
   • Enter the part names and the number of parts on the evaluation form and key in the data to the personal computer in the same order as the attaching sequence.
   • Determine the attaching sequence of the subassembly units.
   • Determine the parts-attaching procedures.
   • Enter the symbols for each part on the evaluation form and key in the data to the personal computer.
3. Calculate Evaluation Indices:
   • Calculate $E$, $K$, and $AT$ by the personal computer.
   • K value: Only able to calculate in redesign.
4. Evaluation Index Judgement:
   • Compare $K$ to the target value (desirable to be below 0.7).
   • It is desirable that $E$ be over 80 points for easier assembly.
5. Improve Design:
   • Prepare proposed improvements:
     • Find subassemblies and parts having relatively small $E$ values, then attempt to reduce the number of parts $N$ and simplify the attaching procedure.
     • A reduction in $N$ sometimes results in a small $E$. In such cases, a reduction in $N$ is preferred to a smaller $E$.
     • When the design is improved, gradual improvements in $E$ (20 to 30 points) are desirable.

Example 1 - Simple Structure

• Spot facing: A finishing operation to produce a flat round surface, usually around a drilled hole, for proper seating of a bolt head or nut; done using a special Spot facing tool.
• The example shows different structure designs (Structure 2, Structure 3) and their impact on assemblability evaluation score ($E$) and assembly cost ratio ($K$).
• Improvements can be seen when the orientation of a part is maintained, or when press-fitting is used in place of screws.

Example 2 - House Supply Fuse Assembly

• Fuse wire enclosed in an “evacuated glass tube with metal end caps”.
• Original double cap design assembly sequence reduced from 20 to 11 in redesign 1, and 12 in redesign 2.
• Illustrates the use of Hitachi's evaluation sheet for the original double cap design.
• Shows how to calculate penalty points, $T$, $\Sigma [T(n)]$, $\Sigma [CFE \times n]$, $M$, PCDE, $AT$, $E$, and $SF$.

Result Interpretations

• Original double cap design:
  • $AT = 20$
  • $E = 40$
  • $PCDE = 0.5$
  • $SF = 20$
  • $N = 8$
  • An $E$ score of 40 indicates a reasonably difficult assembly.
  • $SF$ gives a more reliable impression of design efficiency as it includes the PCDE.
  • $SF$ score of 20 indicates a poor design.
  • $PCDE$ of 0.5 means only half of the parts are fundamentally important.
  • These scores indicate areas for improvement.
  • There are 4 candidates for elimination (CFE) – 2 end caps and 2 card discs.
  • The most difficult part to assemble is the fuse element (but it is a necessary part).
• Single cap redesign 1:
  • $AT = 10.5$
  • $E = 38$
  • $PCDE = 1$
  • $SF = 38$
  • $N = 4$
  • $K = 0.53$
• Single cap redesign 2:
  • $AT = 9.7$
  • $E = 41$
  • $PCDE = 1$
  • $SF = 41$
  • $N = 4$
  • $K = 0.49$
  • The main difference between redesign 1 and 2 accounts for the two ways the fuse element is soldered.
• Both redesigns 1 and 2 have a reduced part count of 4, leading to a PCDE of 1.
• $E$ scores of redesigns 1 and 2 are similar, so we look at $SF$.
• Redesign 2 has the highest $SF$ and smallest $K$ values, therefore is selected.
• $SF$ score of 41 indicates a much improved design.
• $K = 0.49$ means redesign 2 will cost 49% of the original design to assemble.

Contact Details

• DR ARFAUZ A RAHMAN
• Room 06.018, Ashby Building
• Stranmillis Road Belfast
• Northern Ireland BT9 5AH
• Tel: +44 (0) 28 9097 5495
• Email: arfauz.arahman@qub.ac.uk