Specialized Technique #18: Mastering Interference Calculation in Textile Multi-Machine Assignment

In advanced methods engineering, we define interference not as simple dead time, but as a critical stochastic variable determining the profitability boundary between operator cost (Opex) and asset efficiency (OEE).

For the Industrial Engineer or Plant Manager, multi-machine assignment has ceased to be an exercise of intuition or linear timing. In this technical analysis, we break down how to move from deterministic models to the mathematical precision of probabilistic models, avoiding hidden losses that, according to our field data, can reach up to 360 machine-hours/month in poorly balanced standard weaving plants.

The Mathematical Problem of Interference in the Textile Sector

Machine assignment in the textile industry (whether spinning, weaving, or finishing) presents a challenge that simple arithmetic cannot solve: the randomness of stops.

If we apply a linear calculation where theoretical load is $N \times (t/T)$ (with $N$ being the number of machines, $t$ service time, and $T$ cycle time), we are assuming that machines never stop simultaneously. This is a fundamental error. Under Queuing Theory, stops in looms and spinning frames generally follow a Poisson Distribution.

When a machine stops and the operator is busy attending another (or performing patrolling tasks), machine interference time ($I$) is generated.

Difference between Theoretical vs. Real Cycle Time

The common failure in production planning is calculating capacity based on Theoretical Cycle Time. However, reality on the plant floor imposes the following equation:

$\text{Real Cycle} = \text{Machine Time} + \text{Operator Time (Manual)} + \text{Interference ($I$)}$

Ignoring variable $I$ causes overestimation of installed capacity and unreachable production scheduling.

The hidden cost in OEE and operator saturation

According to our sector benchmarks:

• In assignments exceeding 6 machines (especially in high-speed air-jet looms), poor interference management generates losses of 8% to 14% of available time.
• This directly impacts Availability within the OEE calculation, penalizing the ratio without the maintenance department being responsible.

Validated Calculation Methodologies (From Ashcroft to Industry 4.0)

To determine interference with technical rigor, at Cronometras we evaluate three methodological pathways, discarding linear timing due to its inability to capture event simultaneity.

Ashcroft Tables (The Traditional Standard)

The classic method uses the service coefficient ($k$), where $k$ is the ratio between service time and machine running time. Ashcroft tables assume stops are random and repair times are constant or exponential.

• Advantage: Calculation speed for initial estimates.
• Limitation: Loses significant precision if there is high variability in repair times (e.g., complex warp knotting vs. simple bobbin change).

Adapted Wright Formula

Specifically useful in Ring Spinning environments where attention priorities exist. This formula adjusts interference based on time ratios and the number of machines:

$I = 50 \times \left[ \left( (1 + X - N)^2 + 2N \right)^{1/2} - (1 + X - N) \right]$

Where $X$ represents the service time ratio. This model is superior when patrolling is cyclic but interventions are prioritized.

Monte Carlo Simulation (2025 Trend)

Modern methods engineering demands integration with MES systems. The trend for 2025 is using predictive algorithms.

• We feed the model with real MTBF (Mean Time Between Failures) extracted from the machine PLC.
• We simulate thousands of operating scenarios to predict interference with a standard deviation ($\sigma$) lower than 2%.
• This allows adjusting multi-machine assignment dynamically according to raw material quality (which affects breakage rate).

Technical Note: To delve into base data capture, consult our article on Predetermined Time Systems (MTM/MOST), essential for feeding these simulations.

Regulatory Framework and Labor Scenario Spain 2025

Technical calculation must be shielded against audits and works councils. Interference is not just a productive datum; it is a labor variable.

ILO Standards and “Vigilant Attention”

The International Labour Organization (ILO) is clear: interference must be compensated.

1. If the operator waits for a machine to finish its cycle, that time is not rest; it is effective working time (vigilant attention).
2. If the operator works on a machine and others wait (interference), the incentive system must contemplate this production loss beyond the worker’s will.

Saturation Limits and Fatigue (ISO 10075)

Projection of the 2025 labor scenario in Spain suggests stricter control over mental load.

• Recommended Maximum Saturation: 75-85% (ILO/Bedaux scale), including fatigue allowances.
• Cognitive Ergonomics: Interference generates stress. Multiple simultaneous alarms increase mental load. An assignment calculation saturating the operator at 95% theoretically will result in a drop in performance and quality due to human factors under stress.

CRONOMETRAS Solution: Workflow for Optimal Assignment

As a specialized consultancy, we implement a three-phase workflow to solve the assignment equation:

Phase 1: Standardization Via MTM-2 / MOST

You cannot calculate reliable interference on an unstable method. Before modeling, we eliminate human variability by applying MTM-2 or MOST.

• Objective: Standardize intervention time (breakage repair, bobbin change). An operator with an inefficient method exponentially skyrockets interference.

Phase 2: High Precision Work Sampling

We conduct a Work Sampling to categorically differentiate:

1. Technical Stops (Stochastic): Yarn breaks, weft faults.
2. Cyclic Stops (Deterministic): Beam changes, doffing.

Phase 3: Determining Economic Optimum ($N_{opt}$)

The ultimate goal is not maximizing production per operator, but minimizing Total Cost. We use the following cost function:

$\text{Total Cost} = (C_o \times N) + (C_m \times I)$

• $C_o$: Operator hourly cost.
• $C_m$: Machine downtime hourly cost (Profit Loss).
• $N$: Number of assigned machines.

If $C_m$ is high (latest generation machinery with high amortization), the optimal strategy reduces $N$ to minimize $I$. If machinery is fully amortized, we can tolerate higher $I$ to optimize $C_o$. For operating cost reduction strategies, review our section on OEE and Productivity Consulting.

Use Cases: Spinning vs. Weaving vs. Finishing

Interference nature varies by subsector:

1. Spinning (Ring Spinning): Interference is usually linked to spindle breakage frequency and patrolling time. Here Wright’s model is predominant.
2. Weaving: In Jacquard or air-jet looms, knotting complexity varies. Interference here severely impacts fabric quality (stop marks).
3. Finishing: Often more continuous or batch processes, where interference occurs in loading/unloading phases and recipe preparation.

Conclusions for Engineering Management

Machine interference management is the litmus test for Methods Engineering in the textile sector.

• Migration: It is imperative to abandon linear static calculations.
• Dynamism: Any change in machine RPM or yarn quality requires immediate reassignment calculation.
• Holistic Vision: Multi-machine assignment must balance human saturation (Ergonomics/ISO) with asset profitability (OEE).

Is your plant losing capacity due to poorly calculated interference?

An error in the assignment coefficient could be costing thousands of euros in monthly lost profit. Do not base your production on intuition.

REQUEST MULTI-MACHINE ASSIGNMENT DIAGNOSIS

Asset for Engineers: Download the Simplified Interference Excel Calculator (Ashcroft Model) here and start validating your current assignments.

This article is part of the technical series on Industrial Process Audit by Cronometras.