By Chris Monchinski
Reviewed by Nasser Ahmad, Nikhil Joshi
For decades, manufacturers have relied on metrics such as Overall Equipment Effectiveness (OEE), throughput, yield, and cycle time to evaluate performance. These metrics remain valuable, but they share a common limitation: they are primarily lagging indicators. They tell us what happened after production is complete. As manufacturers pursue digital transformation, artificial intelligence, digital twins, and real-time operational intelligence, a different type of metric is becoming increasingly important—one that measures how a process performs while it is running rather than after it is finished. This concept is not new. In fact, MESA introduced it years ago through its Time-in-State (TIS) methodology. Today the principles behind Time-in-State deserve renewed attention, particularly in batch manufacturing industries such as pharmaceuticals, biotechnology, specialty chemicals, food and beverage, and coatings.
The Limitation of Traditional Metrics
Consider a typical batch process. At the end of execution, the organization may evaluate:
- Batch yield
- Batch cycle time
- Right-first-time performance
- Throughput
- Equipment utilization
While these metrics are essential, they often fail to explain why a batch succeeded or failed. A batch may achieve acceptable yield while operating outside optimal conditions for a significant portion of its execution. Conversely, a batch may narrowly miss a target despite operating within desired process conditions nearly the entire time. What manufacturers increasingly need is a metric that quantifies process stability and operational excellence while production is occurring. That is precisely what Time-in-State was designed to provide.
What Is Time-in-State?
At its core, Time-in-State measures the percentage of time a process spends operating within predefined conditions.
Rather than asking: "Did we make good product?" … Time-in-State asks: "How much of the process was executed under the conditions known to produce good product?"
Every manufacturing process can be described by operating states:
- Ideal
- Acceptable
- Degraded
- Critical
- Fault
The objective becomes maximizing the amount of time spent in the ideal state while minimizing time spent in degraded or abnormal conditions. This subtle shift changes the focus from outcome measurement to operational behavior.
Why Time-in-State Now?
Time-in-State provides a powerful framework for driving measurable business values when considering how manufacturing data can be leveraged for more immediate control and improved, consistent quality. Having a disciplined set of KPI provided by time-in-state can be a powerful framework from which to leverage predictive control and AI tools. Consider these key business outcomes that can be driven my time-in-state:
- Reduced batch cycle time through earlier identification of process drift
- Increased asset utilization by minimizing time spent in degraded states
- Improved product consistency and reduced quality deviations
- Faster root cause analysis and continuous improvement efforts
- Better operational context for AI and predictive analytics initiatives
Applying Time-in-State to Batch Manufacturing
The concept is particularly powerful in batch operations because every batch naturally progresses through a sequence of phases. Examples include:
Preparation
- Materials available
- Equipment ready
- Recipe loaded
- Operators available
Charging
- Proper addition sequence
- Target feed rates achieved
- Environmental conditions maintained
Processing
- Temperature within limits
- Pressure within limits
- Agitation within limits
- Process control operating as expected
Hold
- Residence time maintained
- Environmental conditions maintained
Transfer
- Flow rates maintained
- No interruptions
- No quality excursions
For each phase, organizations can define what constitutes an ideal operating state. Rather than simply measuring phase duration, manufacturers can calculate:
- Percentage of time in ideal state
- Percentage of time in acceptable state
- Percentage of time in degraded state
- Percentage of time in critical state
This creates a far richer picture of operational performance than traditional duration-based metrics.
Identifying the Key Influencing Factors
Not every variable matters equally. MESA's methodology emphasizes the identification of Key Influencing Factors (KIFs)—the variables most responsible for driving process outcomes. Examples include:
Biopharmaceutical Cell Culture
- pH
- Dissolved oxygen
- Temperature
- Feed profile
- Viable cell density
Specialty Chemicals
- Reaction temperature
- Feed ratio
- Residence time
- Agitation speed
Lyophilization
- Shelf temperature
- Chamber pressure
- Product temperature
Food Processing
- Moisture content
- Temperature profile
- Cooking time
- Product flow rate
These variables become the foundation of the Time-in-State model. The process state is continuously evaluated based on whether these critical factors remain within their desired operating ranges.
A New Approach to Performance Management
Many organizations continue to rely heavily on OEE as their primary operational metric. OEE works exceptionally well for high-speed discrete manufacturing environments where machine availability is the primary constraint. Batch manufacturing often operates differently. The most important questions frequently become:
- How effectively are reactors utilized?
- How quickly are batches completed?
- How much time is lost to waiting, cleaning, quality adjustments, and rework?
- How often do process variables drift outside optimal ranges?
Time-in-State provides direct visibility into these issues. For example, a process may reveal:
|
State |
Percent Time |
|
Ideal |
72% |
|
Acceptable |
18% |
|
Degraded |
8% |
|
Critical |
2% |
These measurements immediately identify improvement opportunities that traditional yield or cycle-time metrics may conceal.
Time-in-State and ISA-95
One of the more interesting aspects of the Time-in-State methodology is its alignment with ISA-95. Time-in-State metrics can be calculated across multiple levels of the enterprise:
- Equipment
- Unit
- Production Line
- Area
- Site
- Enterprise
This aligns naturally with ISA-95 equipment and operations models. As organizations move toward Unified Namespace architectures, manufacturing knowledge graphs, and data-centric operations, Time-in-State can become a standardized performance attribute attached to equipment, processes, products, and operations. Rather than simply publishing production counts or machine states, manufacturers can publish operational effectiveness in near real time.
Time-in-State and the Digital Twin
Digital twins seek to answer a fundamental question: "How closely does actual performance match intended performance?" Time-in-State offers a practical way to measure this alignment. A validated process model effectively defines the ideal state. Actual execution can then be continuously compared against that model. The resulting Time-in-State metrics become indicators of process conformance. In regulated industries, this provides a powerful bridge between process validation, continued process verification, and operational excellence.
Time-in-State as an AI Metric
Perhaps the most exciting application of Time-in-State is its role in artificial intelligence.
AI systems require objective measures of success. Historically, manufacturers have attempted to optimize:
- Yield
- Throughput
- Scrap
- Downtime
While useful, these metrics often arrive too late to support proactive decision-making. Time-in-State creates a real-time measure of process health. High-performing batches become examples of high Time-in-State execution. Lower-performing batches become examples of low Time-in-State execution. This creates a natural training signal for AI models, digital twins, and operational copilots. Rather than asking AI to maximize yield directly, organizations can ask:
"What actions will maximize Time-in-State?" The resulting recommendations are often more actionable and easier to explain to operators and process engineers.
Looking Forward
As manufacturers continue their digital transformation journeys, there is growing recognition that outcome metrics alone are insufficient. Organizations need metrics that measure how work is being performed, not simply what was produced. Time-in-State provides exactly that capability. It bridges operational excellence, process control, digital twins, artificial intelligence and Smart Manufacturing initiatives into a single framework focused on process stability and operational performance. MESA recognized the value of this concept years ago. Today, as manufacturers seek more predictive and intelligent operations, Time-in-State may be more relevant than ever.
References
MESA WhitePaper 47 - Time-in-State Metrics
MESA WhitePaper 48 - Time-in-State Metric Implementation Methodology
MESA WhitePaper 50 - Time-in-State Metrics #3 2014-6
