A New Method to Calculate Industrial Robot Standard Time
In the fast-evolving landscape of modern manufacturing, where automation and intelligent systems are rapidly redefining production lines, a critical challenge has emerged: how to accurately measure productivity in environments dominated by industrial robots. Traditional methods for calculating standard time—the foundational metric for performance and efficiency in manufacturing—are rooted in human labor models. These models account for worker fatigue, skill variation, and personal breaks. But when humans are replaced by machines, these factors no longer apply. A new framework is needed, one that reflects the unique operational characteristics of industrial robots, particularly their susceptibility to mechanical and electronic failures.
A groundbreaking study published in Modern Manufacturing Engineering proposes a novel approach to this challenge. Led by ZHANG Yuxian, LIN Jing, YUE Fang, and WANG Bing from the Business School at Guilin University of Electronic Technology, the research introduces a comprehensive method for calculating the standard time of industrial robots in intelligent manufacturing systems. The work not only redefines the components of standard time in a robotic context but also integrates probability theory to quantify the impact of machine failures—an often-overlooked yet critical factor in real-world production environments.
The shift toward smart manufacturing, accelerated by national strategies such as China’s “Made in China 2025,” has led to widespread deployment of industrial robots across assembly lines. These machines perform repetitive tasks with high precision and consistency, theoretically enabling uninterrupted production. However, in practice, no machine operates flawlessly. Electronic malfunctions, sensor errors, mechanical wear, and software glitches all contribute to unplanned downtime. These interruptions, while not part of the robot’s programmed work cycle, directly affect overall productivity. Ignoring them in standard time calculations leads to overly optimistic performance benchmarks and flawed capacity planning.
The research team recognized this gap and set out to develop a model that reflects the reality of robotic operations. Their approach begins with a fundamental re-conceptualization of what constitutes standard time in a robotic system. Unlike traditional models, which focus on human operators and include allowances for rest and personal time, the new model is built around three core components: failure-free time, failure time, and management time.
Failure-free time represents the ideal operational duration—the time a robot spends executing its programmed tasks without interruption. This is the closest equivalent to the “normal time” in human-based time studies. For industrial robots, this value is highly deterministic. Once a robot’s motion path, speed, and task sequence are programmed, the time required to complete a cycle is fixed. In the study, this value can be derived directly from the robot’s control parameters or measured empirically using time study techniques, such as stopwatch analysis, during stable operation.
The second component, failure time, addresses the Achilles’ heel of any mechanical system: breakdowns. While robots do not tire or require breaks, they are prone to failures that halt production. The researchers argue that these failures must be systematically accounted for in standard time calculations. To do so, they introduce a probabilistic model that quantifies the average time lost due to failures across a given production period.
The model classifies failures into distinct types—such as sensor malfunctions, motor failures, or communication errors—each with its own probability of occurrence and associated downtime. By analyzing historical maintenance logs and operational data, the team assigns a failure probability and repair duration to each type. These values are then used to compute an expected downtime per production cycle, effectively averaging the impact of random failures over time.
A key innovation in the methodology is the use of fault grouping to simplify analysis. With dozens of potential failure modes, calculating each one individually would be impractical. Instead, the researchers propose clustering failures with similar downtime durations into groups, using a relative error threshold to ensure consistency. This grouping reduces complexity while preserving accuracy, allowing for a more manageable and scalable calculation process.
The third component, management time, accounts for planned interruptions that are not part of the robot’s core task. These include scheduled maintenance, software updates, calibration procedures, and system reconfigurations for product changeovers. Unlike human workers, robots do not require personal or fatigue allowances. However, they do require administrative oversight and periodic servicing. The study proposes a “management allowance rate,” typically set between 3% and 5%, to incorporate these planned downtimes into the standard time.
By combining these three elements—failure-free time, expected failure time, and management allowance—the researchers derive a comprehensive formula for standard time. The model first calculates a “normal time” by adding failure-free time and average failure time. This normal time is then adjusted upward by the management allowance rate to arrive at the final standard time. This two-step process mirrors the traditional “rating and allowance” method used in human time studies but adapts it to the mechanical and probabilistic nature of robotic systems.
To validate their model, the team applied it to a real-world case: an automated rubber plug feeding robot in a manufacturing plant. The robot was responsible for transporting and positioning rubber components for stamping, completing one cycle every four seconds under ideal conditions. Over a six-month period, the robot operated for 2,640 hours and produced 2,262,400 units. During this time, 13 distinct types of failures were recorded, ranging from minor sensor errors to major motor failures, with downtime varying from a few minutes to several hours.
Using the historical failure data, the researchers calculated the probability of each failure type based on the proportion of total operational time it consumed. They then grouped similar failures and computed the expected downtime per cycle. Applying a 3% management allowance, they arrived at a standard time of 4.12021 seconds per unit—slightly higher than the ideal 4-second cycle.
This seemingly small adjustment has significant implications. When used to project annual capacity, the model predicted a theoretical output of approximately 2,306,678 units, which was within 2% of the actual production volume. This close alignment demonstrates the model’s accuracy and practical utility. In contrast, using the ideal 4-second cycle would have overestimated capacity by more than 44,000 units, leading to potential overcommitment of resources and unmet delivery targets.
The implications of this research extend beyond a single factory floor. As global manufacturing continues its shift toward Industry 4.0, the ability to accurately measure and predict robotic productivity becomes increasingly vital. Traditional key performance indicators (KPIs) such as Overall Equipment Effectiveness (OEE) already incorporate availability, performance, and quality. However, standard time remains a foundational input for labor and capacity planning, cost accounting, and scheduling systems.
By providing a scientifically grounded method for calculating robotic standard time, this study fills a critical gap in the toolkit of industrial engineers and operations managers. It enables more realistic production planning, better resource allocation, and more accurate cost modeling. Moreover, the probabilistic approach allows manufacturers to simulate the impact of different maintenance strategies or reliability improvements on overall productivity.
For example, if a company is considering upgrading its robot’s sensors to reduce detection errors, the model can quantify how much time—and therefore money—would be saved by reducing the frequency of a specific failure type. This transforms maintenance from a reactive cost center into a strategic lever for performance optimization.
The research also has broader implications for the future of work in manufacturing. As robots take over more tasks, the role of human workers is shifting from direct operation to supervision, maintenance, and system optimization. Accurate standard time calculations for robotic processes allow companies to rebalance workloads, retrain employees, and redesign workflows in a data-driven manner.
Furthermore, the model’s reliance on historical data and probability theory makes it adaptable to different types of robots and production environments. Whether in automotive assembly, electronics manufacturing, or pharmaceutical packaging, the core principles remain applicable. The only requirements are reliable records of failures and operational time—data that most modern smart factories already collect through their industrial IoT systems.
One of the strengths of the study is its grounding in established industrial engineering principles. Rather than discarding decades of time study methodology, the authors build upon it, reinterpreting its components for a new technological era. This continuity ensures that the model can be integrated into existing enterprise resource planning (ERP) and manufacturing execution systems (MES) without requiring a complete overhaul of current practices.
The use of probability theory to model failure impact is particularly noteworthy. In traditional manufacturing, variability in human performance is often treated as noise to be minimized. In contrast, the new model treats machine failure as a quantifiable risk—a predictable statistical phenomenon rather than an unpredictable disruption. This shift in perspective aligns with modern risk management practices in finance, logistics, and supply chain management, bringing a more sophisticated analytical lens to production engineering.
The research also highlights the importance of data quality in smart manufacturing. The accuracy of the standard time calculation depends heavily on the completeness and reliability of failure records. Sporadic or poorly documented downtime events can skew the probability estimates and lead to inaccurate models. This underscores the need for robust data governance and real-time monitoring systems in automated factories.
Looking ahead, the model could be extended in several directions. One possibility is the integration of predictive maintenance data. If machine learning algorithms can forecast the likelihood of a failure based on sensor readings, these predictions could be incorporated into the standard time model, making it dynamic rather than static. A robot’s standard time could then vary based on its current health status, enabling real-time adjustments to production schedules.
Another potential extension is the inclusion of energy consumption or environmental impact as a factor in standard time. As sustainability becomes a key concern for manufacturers, a more holistic definition of productivity may emerge—one that balances output, time, cost, and carbon footprint. The modular structure of the current model makes it well-suited for such expansions.
The work also raises philosophical questions about the nature of work and efficiency in an automated world. If robots can operate nearly 24/7 with minimal downtime, does the concept of “standard time” still hold the same meaning? Or does it become a measure not of human capability, but of machine reliability? The researchers’ answer seems to be that standard time remains relevant, but its definition must evolve to reflect the new realities of production.
In an era where artificial intelligence is beginning to control complex manufacturing systems, the need for transparent, interpretable models is greater than ever. Black-box algorithms may optimize performance, but engineers and managers need to understand why a system behaves a certain way. The model proposed by ZHANG Yuxian and colleagues offers just that: a clear, logical, and mathematically sound framework that bridges the gap between theory and practice.
The study’s practical validation in a real production environment adds significant credibility. Too often, academic research remains confined to simulations or controlled lab settings. By applying their method to actual operational data and demonstrating its predictive accuracy, the team has shown that their approach is not just theoretically sound but also operationally viable.
As smart factories become the norm rather than the exception, the tools used to manage them must keep pace. This research represents a crucial step in that direction. It provides a much-needed method for measuring productivity in robotic systems, one that accounts for the inevitable imperfections of real-world machinery. By doing so, it helps ensure that the promise of intelligent manufacturing—greater efficiency, lower costs, and higher quality—can be realized in practice, not just in theory.
The implications of this work are likely to resonate across the global manufacturing sector. From small workshops adopting their first robot to multinational corporations running fully automated plants, the ability to accurately calculate standard time will be essential for competitiveness. As robotics technology continues to advance, so too must the methods we use to understand and optimize their performance.
In conclusion, the research by ZHANG Yuxian, LIN Jing, YUE Fang, and WANG Bing offers a timely and valuable contribution to the field of industrial engineering. By redefining standard time for the age of robotics, they have provided a practical, data-driven tool that can help manufacturers navigate the complexities of intelligent production systems. Their work stands as a testament to the enduring relevance of industrial engineering principles, even as the tools and technologies of manufacturing undergo radical transformation.
ZHANG Yuxian, LIN Jing, YUE Fang, WANG Bing, Business School, Guilin University of Electronic Technology, Modern Manufacturing Engineering, DOI: 10.16731/j.cnki.1671-3133.2021.12.016