Kanban System and Industrial Time Study: Synchronizing Materials and Standard Times

Kanban System and Industrial Time Study: Synchronizing Materials and Standard Times

• Cronometras Team

Learn how to integrate industrial time study with the Kanban system: takt time, Kanban card calculation, buffer and supermarket sizing, and digital tools to keep cycle times up to date.

At the heart of lean manufacturing lies a non-negotiable premise: produce only what is needed, when it is needed, and in the exact quantity. The Kanban System, popularized by Toyota in the 1950s, is the signaling mechanism that puts this just-in-time philosophy into practice. But there is a detail that many overlook: without accurate standard times, Kanban is like a GPS without coordinates. Cards, buffers and supermarkets depend entirely on knowing how long each operation really takes. This is where industrial time study becomes the silent —and indispensable— ally of every pull system.

In this article we will explore how to integrate time study with the Kanban methodology: from calculating takt time to sizing cards and buffers, through practical cases with formulas and the use of digital tools such as CronometrasApp to keep data updated in real time.


Fundamentals of the Kanban System: Produce What Is Consumed

Kanban —a Japanese word meaning "card" or "signal"— is a production control system that uses visual signals to authorize the manufacturing or movement of materials. Unlike traditional push systems (where each station produces according to a centralized plan and "pushes" the product to the next stage), Kanban implements a pull flow: customer demand —internal or external— triggers production.

The key elements of the system are:

  1. Production Kanban card: authorizes the manufacturing of a defined quantity of parts.
  2. Transport Kanban card: authorizes the movement of material between stations.
  3. Kanban supermarket: controlled storage point between processes.
  4. Kanban board: visualization of the status of the cards.

The principle is simple: when a customer (the next station) consumes products from the supermarket, the Kanban card returns to the supplying process as authorization to replenish exactly what has been consumed. But how do we know how many cards we need and how much each one should contain? The answer lies in the actual cycle times, obtained through industrial time study.


Takt Time: The Pulse of Production

The takt time is the rate at which one unit must be produced to meet customer demand. It is calculated with a surprisingly simple formula, but one with far-reaching implications:

Takt Time=Available Production TimeCustomer Demand\text{Takt Time} = \frac{\text{Available Production Time}}{\text{Customer Demand}}

For example, if a plant operates 480 minutes per shift (8 hours) with 40 minutes of breaks and meetings, the net available time is 440 minutes. If daily demand is 880 units:

Takt Time=440 min880 units=0.5 min/unit=30 seconds\text{Takt Time} = \frac{440 \text{ min}}{880 \text{ units}} = 0.5 \text{ min/unit} = 30 \text{ seconds}

This means that, to satisfy the customer, each station must complete one unit every 30 seconds. If the actual cycle time of an operation —measured by time study— exceeds the takt time, we have a bottleneck. If it is lower, there is idle capacity. In both cases, the Kanban system must be adjusted.

Industrial time study is the only reliable way to obtain the actual cycle time. Estimates, unattributed historical data or theoretical machine times rarely reflect the reality of the gemba.


Calculating the Number of Kanban Cards: The Fundamental Formula

The classic formula for determining how many Kanban cards a loop between two stations needs is:

N=D×Te×(1+α)CN = \frac{D \times T_e \times (1 + \alpha)}{C}

Where:

  • NN = Number of Kanban cards required
  • DD = Demand per unit of time (units/minute, units/hour)
  • TeT_e = Loop waiting time (replenishment lead time): includes processing, queue, transport and waiting time
  • α\alpha = Safety factor (typically between 0.05 and 0.15)
  • CC = Kanban container or lot capacity (units per card)

Breaking Down TeT_e: The Role of Time Study

The waiting time TeT_e is the most critical parameter and the one that requires the greatest precision. It consists of:

Te=Tproc+Tqueue+Ttrans+TwaitT_e = T_{\text{proc}} + T_{\text{queue}} + T_{\text{trans}} + T_{\text{wait}}

Each of these sub-times must be measured with industrial time study:

Component Description Measurement method
TprocT_{\text{proc}} Processing time at the supplying station Cumulative or reset (snapback) time study, with at least 10 observations
TqueueT_{\text{queue}} Waiting time in queue before being processed Work sampling (work sampling)
TtransT_{\text{trans}} Transport time between stations Time study of routes, averaging multiple trips
TwaitT_{\text{wait}} Waiting for a complete lot or resource availability Direct observation and recording

Numerical Example

Let us consider a plastics injection molding process that feeds an assembly line:

  • Demand: D=120D = 120 units/hour =2= 2 units/minute
  • Tproc=4.2T_{\text{proc}} = 4.2 min (average of 15 timed observations, including setup and cleaning)
  • Tqueue=1.8T_{\text{queue}} = 1.8 min
  • Ttrans=0.5T_{\text{trans}} = 0.5 min
  • Twait=0.3T_{\text{wait}} = 0.3 min
  • Te=4.2+1.8+0.5+0.3=6.8T_e = 4.2 + 1.8 + 0.5 + 0.3 = 6.8 min
  • Safety factor: α=0.10\alpha = 0.10
  • Container capacity: C=50C = 50 units

N=2×6.8×(1+0.10)50=2×6.8×1.1050=14.9650=0.2992N = \frac{2 \times 6.8 \times (1 + 0.10)}{50} = \frac{2 \times 6.8 \times 1.10}{50} = \frac{14.96}{50} = 0.2992

Rounding up: N=1N = 1 Kanban card. In practice, this means that 1 container of 50 parts is enough to cover demand during the 6.8-minute replenishment time, with a 10% safety margin.

However, if the processing time had been estimated "by eye" at 3 minutes instead of the actual 4.2 minutes obtained through time study, the calculation would have given N=0.25N = 0.25, incorrectly suggesting that 1 card was enough —when in reality the supplier would take almost 7 minutes to replenish, causing stoppages on the assembly line.


Time Study to Determine Actual Cycle Times: Beyond the Stopwatch

Modern industrial time study has evolved far beyond the mechanical stopwatch and the paper clipboard. Tools such as CronometrasApp make it possible to:

1. Precision Time Study with Elements

Breaking each operation down into measurable elements makes it possible to identify exactly where the time is consumed. For Kanban, this is crucial because:

  • Value-added elements determine the theoretical minimum cycle time
  • Non-value-added elements (walking, searching for tools, adjustments) are improvement opportunities that can reduce TeT_e and, therefore, the number of cards required
  • Customizable element libraries make it possible to standardize measurements across different analysts and shifts

2. Recording Variability

A single average value is not enough. The Kanban system is sensitive to cycle time variability. If a process has an average time of 4.2 minutes but a standard deviation of 2.1 minutes, the safety factor α\alpha must be higher.

αadjusted=αbase+σxˉ\alpha_{\text{adjusted}} = \alpha_{\text{base}} + \frac{\sigma}{\bar{x}}

Where σ\sigma is the standard deviation and xˉ\bar{x} the mean of the observed times. This coefficient of variation reflects the instability of the process and must feed directly into Kanban sizing.

3. Voice Input and Offline Mode

In shop-floor environments where hands are busy and connectivity is intermittent, the voice input and offline mode (PWA) of CronometrasApp make it possible to capture data without interrupting the workflow, synchronizing automatically when the connection is restored.


Sizing Buffers and Supermarkets: How Much Stock Is Enough

The Kanban supermarket acts as a buffer between processes running at different rates. Sizing it correctly is one of the most important decisions —and the one most dependent on time study data— when designing a pull system.

Time-Based Safety Stock

SS=D×σTe×zSS = D \times \sigma_{T_e} \times z

Where:

  • SSSS = Safety stock (in units)
  • DD = Demand per unit of time
  • σTe\sigma_{T_e} = Standard deviation of the loop waiting time (obtained by time study)
  • zz = Service factor (1.65 for 95% service level, 2.33 for 99%)

If in our previous example the standard deviation of TeT_e is 1.4 minutes (measured over 20 replenishment cycles):

SS=2 units/min×1.4 min×1.65=4.62≈5 unitsSS = 2 \text{ units/min} \times 1.4 \text{ min} \times 1.65 = 4.62 \approx 5 \text{ units}

This adds 5 units to the base level of the supermarket, raising the total target stock to 50+5=5550 + 5 = 55 units.

Visual Reorder Point

The traditional Kanban system uses visual marks in the supermarket. When the inventory level drops below a painted line (the reorder point), the replenishment signal is triggered. This point is calculated as:

Reorder Point=D×Te+SS\text{Reorder Point} = D \times T_e + SS

For our case: RP=2×6.8+5=18.6≈19\text{RP} = 2 \times 6.8 + 5 = 18.6 \approx 19 units. When the supermarket drops below 19 parts, the Kanban card must be sent to the supplying process.


Synchronizing Lines with Standard Times

A balanced production line is one where each station operates at a rate equal to or very close to the takt time. Time study provides the data to:

1. Identify the Pacemaker Process

The pacemaker is the process that sets the pace for the entire line. It is usually the station closest to the customer or the one with the longest cycle time. Once identified through time study, all Kanban loops are designed to feed this process without interruption.

2. Adjust the Quantity per Card

The container capacity CC is not arbitrary. It must be chosen so that:

C≈D×TeC \approx D \times T_e

This way, the number of cards NN approaches 1, simplifying visual management. If CC is too small, we will need many cards (administrative complexity). If it is too large, work-in-process inventory (WIP) will increase.

3. Detect Drift Over Time

Cycle times are not static. Tool wear, operator fatigue, lot changes and seasonal variations modify the actual times. A periodic time study program —ideally integrated into the daily or weekly routine with tools like CronometrasApp— makes it possible to detect these drifts before they desynchronize the Kanban system.


Practical Case: Kanban Redesign on an Electronic Assembly Line

Context: A company manufactures electronic controllers. The line has 4 stations: SMD insertion, wave soldering, manual assembly and functional testing. The current system uses Kanban with 3 cards between each pair of stations, but there are frequent stoppages due to material shortages.

Step 1: Time Study

Time studies are carried out with CronometrasApp over 3 shifts, measuring:

Station xˉ\bar{x} (min) σ\sigma (min) Variation Coef.
SMD 2.1 0.3 0.14
Soldering 3.8 1.2 0.32
Assembly 5.2 0.8 0.15
Testing 2.9 0.5 0.17

Step 2: Takt Time Calculation

Daily demand: 400 units. Available time: 450 minutes.

Takt Time=450400=1.125 min/unit\text{Takt Time} = \frac{450}{400} = 1.125 \text{ min/unit}

Warning: All 4 stations exceed the takt time! The line cannot meet demand with the current configuration. This explains the stoppages: the Kanban buffers empty faster than they are replenished.

Step 3: Rebalancing and Kanban Resizing

The solution is to duplicate the assembly station (the slowest one) and adjust the Kanban loops. With two assembly stations in parallel:

Effective assembly time=5.22=2.6 min/unit\text{Effective assembly time} = \frac{5.2}{2} = 2.6 \text{ min/unit}

Still above 1.125, but combined with method improvements identified in the study (elimination of unnecessary travel), it is brought down to 1.1 min/unit.

New Kanban loops recalculated with the formula N=D×Te×(1+α)CN = \frac{D \times T_e \times (1 + \alpha)}{C}:

  • Soldering → Assembly loop: Te=6.2T_e = 6.2 min, C=20C = 20, α=0.12\alpha = 0.12 → N=1.74≈2N = 1.74 \approx 2 cards
  • Assembly → Testing loop: Te=3.1T_e = 3.1 min, C=10C = 10, α=0.10\alpha = 0.10 → N=0.76≈1N = 0.76 \approx 1 card

Result

After implementing the changes, line stoppages were reduced by 73% and WIP decreased by 28%. Time study not only sized the Kanbans correctly, but also revealed the root cause of the problem: an unbalanced line that no card system could compensate for.


Digital Integration: CronometrasApp as a Single Source of Truth

Modern Kanban management increasingly relies on digital systems (e-Kanban), but the weakest link is still the quality of the input data: the actual cycle time. CronometrasApp closes this gap by offering:

  • Dual stopwatch (reset / cumulative) to adapt to any measurement methodology
  • Customizable element libraries that standardize the breakdown of operations across analysts
  • Automatic calculation of OIT allowances to obtain standard times from observed times
  • Export to Excel and PDF to share data with continuous improvement teams and the ERP/MRP systems that manage Kanban calculations
  • Real-time collaborative work so that multiple analysts can cover different stations simultaneously

Common Mistakes When Integrating Kanban and Time Study

  1. Using historical times without verifying them: times recorded 6 months ago probably do not reflect current reality. Time study must be a living activity.
  2. Ignoring variability: using only averages without considering the standard deviation leads to frequent stockouts.
  3. Not updating the Kanbans after process improvements: if a Kaizen event reduces cycle time by 20%, the Kanban loops must be recalculated immediately —otherwise unnecessary inventory will be maintained.
  4. Confusing cycle time with takt time: cycle time is measured with a stopwatch; takt time is calculated from demand. They are different concepts that must be compared, not confused.
  5. Oversizing out of fear of stoppages: too many Kanban cards inflate inventory and hide problems. A healthy Kanban system must have a certain "tension" that forces continuous improvement.

Conclusion: Kanban Without Time Study Is a Gamble

The Kanban System is one of the most elegant and powerful tools in lean manufacturing, but its effectiveness is conditioned by the quality of the data that feeds it. Without accurate, up-to-date and statistically representative cycle times, any Kanban calculation is worthless.

Industrial time study —boosted by digital tools like CronometrasApp— provides that objective foundation. It makes it possible to calculate the correct takt time, size cards and buffers precisely, detect imbalances before they cause stoppages, and keep the system synchronized over time.

In the era of Industry 4.0, where data flows in real time between sensors, ERPs and dashboards, time study remains —as it was in Toyota's workshops 70 years ago— the primary source of truth on the ground. Because no digital transformation can improve what has not first been measured.

Request a Cronometras Demo


Want to go deeper into integrating time studies with lean systems? Explore our articles on Value Stream Mapping, Line Balancing and OEE Calculation.


Don't want to do it yourself? We do it for you from €800

If you prefer to outsource it, Cronometras delivers a complete time and motion study from a single video: elemental breakdown, OIT/TAL fatigue allowances, standard time with tolerances, and a report in PDF + Excel + JSON ready to present to management.

👉 Request the study from €800 · More about Cronometras

Did you find this useful?

Share this article with your professional network