Operational Excellence is high on the agenda at many companies. Production process optimization presents new challenges, but it also offers new opportunities. The following article describes the Bayer Technology Services approach.
Global competition is still in its infancy, and we have no idea how difficult the market will eventually become. At the last VDE Congress, EU Commissioner Günter Verheugen zeroed in on the key issue. He pointed out that the challenge facing us is to maintain the position which we had in the year 2000. It is futile to try to compete with low labor costs if we want to protect our standard of living. We will have to rely on innovative, intelligent technologies to create competitive production processes which will help us retain our manufacturing base in Europe. This was obviously a direct challenge to engineers to take control of their own fate. If you look at the actual manufacturing process, labor costs in the process industry normally account for only a relatively small proportion of production costs (in the region of 10%). Operational excellence is the term which is used to describe efforts to reduce labor costs and improve the performance of the production process.
The aim is to “eliminate loss and waste in production”, or perhaps “bring production up to the optimum level” is the better wording. Process management makes an important contribution to production efficiency by significantly enhancing process performance (equipment utilization, yield, product quality, availability, safety and delivery performance) and reducing cost (energy consumption, raw material consumption, inventory levels, human resources and capital resources).
Bayer Technology Services (BTS) uses process management tools to improve the entire operational process, from production right through to logistics.
BTS can also optimize subprocesses on existing production systems. OpX optimization offers significant potential, and project experience has shown that state-of-the-art process management techniques alone can produce the following results:
- 8% reduction in energy consumption
- 10% higher throughput
- 15% reduction in inventory
- 70% cost reduction for lab analysis
- 80% reduction in time to product release
- 30% fewer off-spec goods
- 20% reduction in time to production system start-up
- 5% higher system availability
- 90% fewer production planning alerts
- 10% fewer batches per production order
Only a moderate investment is needed to achieve savings that can be as high as 10% of production costs. The expected payback period is 2–12 months.
Based on three main elements
Back in 2004, ARC stated that operational excellence is based on three main elements: performance intelligence, performance management and performance enablement . To achieve operational excellence, BTS prefers the six-sigma-based DMAIC approach (define & measure, analyze & improve, control) on OpX projects. DMAIC was described in detail by Friedrich back in 2005 . According to Friedrich, the current state of production is assessed during an initial optimization potential analysis (define & measure).
The focus during this phase is on the major cost pools and identification of the key performance indicators (KPI) which affect production efficiency. The next step is to define the content and scope of the optimization project. During the second phase (analyze & improve), the team looks into the cause and effect relationships and develops the action plan, project plan and budget for the final implementation phase. In the third phase (control), it is essential for the success of the entire project that the solutions which have been identified are adapted to the existing systems and workflows, the workers are brought on board and trained early on and monitoring and control functions are implemented. There is no other way to ensure long-term success.
Process control technology
Process optimization is still based on process control technology, but the role of online analysis continues to increase. These two elements define the useable information set which can be acquired from the process. Process control systems can use actuators to intervene in the process. Without this capability, even the most intelligent OpX IT systems are completely useless in a production environment. The latest advances in technology present a whole series of challenges to the engineering team. The days when their main task was to control pressure, temperature and flow should now be long gone. The goal today is not to get the process up and running, but rather to focus on overall performance.
The task of process control engineers goes beyond selection of the right instrumentation. They also have to evaluate a whole range of technical parameters and then work together with colleagues from advanced process control to determine the best sensor placement based on dynamic modeling. No manufacturing team is interested in column temperatures. What they really want is to optimize column efficiency and energy consumption. It is equally important to recognize when temperature measurement is no longer sufficient to manage the process, as is shown by the aromatics predistillation example in Fig. 2. Methods like concentration measurement must be used in this application, and placement is critical. As is clearly evident from Fig. 3, process management will not lead to operational excellence without process models which accurately reflect production and logistics flows. These models turn data into information and help exploit the maximum potential of the process . Data acquisition, which involves continuous evaluation and interpretation of the data, plays a key role. A statistical toolkit and a comparison with rigorous process models are a big help. Operational excellent only makes sense if you have a team of well-qualified and experienced engineers who are able to use this approach effectively.
When you have all of this in place, you are nearly there. The next logical step is to use performance monitoring to optimize existing processes and systems, and the BTS PerMonDo product offers this functionality. Continuous monitoring of process and production system benchmarks, which are described in the key indicators, can help enhance process and product system performance and reduce manufacturing costs. Typical key performance indicators are: energy consumption (e.g. solvent concentrations, steam consumption at a specific step in the process); product quality (concentration of chemical components, contamination, secondary component residues); production system utilization (throughput, yield, changeover, retrofit); availability (deposits, blockages, malfunctions, alarms, shut-downs). It is important to assess the key performance indicators in relation to the theoretically possible process state, and rigorous process models are normally needed to do this. Once this assessment capability is in place, performance monitoring can help the engineering team to run the process in a nearly optimal state. The real challenge is to eliminate manual intervention and automate the process based on the key performance indicators. In today’s complex world, for example when large production systems with multiple process steps are started up or shut down, this level of automation is not always easy to achieve. That makes it all the more important to assess the benefits early on using DMAIC analysis.
Process analysis technology
Brief reference was made above to the fundamental importance of real time product quality monitoring in the operational excellence context. Right up into the 1990s, process analysis was focused on measurement outside of the production process to ensure environmental protection and occupation safety. In the meantime, the importance of quality-based process control as a key performance indicator has become equally important. Versatile BTS online process analysis systems are able to acquire virtually any product data. The list includes SpectroBAY, a spectrometer system for process applications, which is available in NIR, MIR, UV and Raman versions, and the BaychroMAT chromatograph which supports the full range of chromatography procedures. All of the systems are fully automated and are supplied with all of the standard process interfaces, a monitoring system and remote diagnosis features. The special Analyzer Results Transfer Software presents the quality data on the process control system in a familiar format, namely as a process control point. BaychroMAT Bio HPLC and BaychroMAT CellCount are innovative online bioanalytics products. Using these tools, it is possible to deploy fully automated fermentation process control based on cell count and protein concentration. This process equipment has a sampling valve, which can be steam sterilized, as well as proven automation components.
What has been discussed for far applies equally to batch and continuous processing. In actual practice, most of the solutions have been deployed on continuous production systems, because energy and raw materials costs are a major competitive factor. A third cost factor becomes more prominent in batch production, namely utilization. This is where Manufacturing Execution Systems (MES) come in.
Manufacturing Execution Systems
Although the documentation and integrated production optimization functions go well beyond detailed production planning, we will concentrate on the planning aspect in the following discussion. One of the most complex MES systems ever deployed went into operation in 2000 at Bayer Crop Science. It is used in a multi-purpose production environment to link highly flexible production modules to production lines which produce specialty products in relatively short runs . About 200 production modules are available, and up to 40 of them can be combined to form a production line. Given this level of flexibility, the main objective of the MES, which is linked to the process control and Enterprise Resource Planning (ERP) systems, was to organize production. The MES identifies production runs and electronically allocates a production order, a recipe and a batch, automatically assembles manufacturer’s instructions and other production documentation, archives them with the batch records and performs module scheduling and coordination. This facility itself is manifesting operational excellence, already (see Fig. 5).
Logistic Execution Systems (LES) needed to be integrated into the workflows on the multi-purpose lines, but BTS has gone one step further in the direction of operational excellence on the MES/LES projects. The focus was firmly on maximization of asset utilization and inventory (working capital) reduction. Integration of MES, LES and supply chain optimization were needed to achieve these goals. BTS consolidated all production planning activities and embedded them in the supply chain management and logistics functions.
The first step is material flow simulation, and it takes place back at the conceptual stage. It looks at the ideal combination of process and logistical systems. Similar models are used today to support production planning and to determine the optimal allocation of production orders, which are normally imported from the ERP systems, to available production capacity. Optimal in the context means that inventory levels as well as asset scheduling are optimized. The best tradeoff between inventory levels, production lot sizes, changeover costs and asset utilization is defined using operational, near-real-time optimization tools. The optimization runs usually take place once a day and cover a time horizon of two to four weeks.
The resulting production plan maximizes the manufacturer’s profit. BTS Evoplan software, which is based on evolution algorithms, is used in a complex two-stage production environment, where first-stage products as well as blends from the second stage are sold. Evoplan helped reduce the number of production orders that could not be produced on time on average from 30 per day to three per day, and the planning horizon was extended from two weeks to four.
Supply chain optimization
Production planning at this level of complexity contributes to supply chain optimization, because the projects address inventory optimization, lot size optimization as a function of product changeover cost and inventory levels, raw material availability, distribution capacity and much more. When it looks at optimization, BTS distinguishes between inventory in transit, strategic inventory to cover expected sales peaks, safety stock to cover demand and production uncertainty and lot-size stock which is used to optimize production lots. BTS uses various models and tools which were developed in house to optimize all four types of inventory, often simultaneously. The main key performance indicator is market service level. Following customizing for the production and logistics processes, the operational tools are made available to customers (see Fig. 6).
In addition to inventory and production optimization, the portfolio includes classic supply chain optimization as well as management functions. This helps to optimize material flows, increase the reliability of requirements planning and generate rough production plans. BTS normally uses standard software which is tailored to the customer’s workflows. The software is provided as operational tools which are rolled out worldwide.
What is missing?
One important aspect of operation excellence has not been addressed yet, namely asset management. At NAMUR, this term means the monitoring of all technical equipment at a production plant, and it goes well beyond standard asset management for field devices . All suppliers still have a lot of work to do in this area. Along with other NAMUR members, BTS can visualize what this type of system might look like. However, a significant amount of development work will be necessary before the PUMA System (Process Unit Monitoring and Alarming) will be making a contribution to operational excellence .
In summary, operational excellence depends on the availability of information about the entire production process including logistics and the deployment of automated systems which intervene in the process. This is a significant integration challenge, but the most difficult tasks are actually to identify the KPI, which are the factors that have the biggest impact on competitiveness, develop the models and mirror the models in ongoing production.
To do this, you need a team of experienced, qualified engineers from various departments who are very familiar with the production, logistics and information flows. Without a highly committed and qualified team, operational excellence cannot be fully implemented and achieved.
 ARC Insight 2004-48 MHLP, Tom Fiske, Dave Woll: Improving Business Performance by Unifiying Operations
 Martin Friedrich, “Optimierung der Produktion – Methoden, Werkzeuge und Vorgehensweisen zur Erschließung versteckter Effizienzreserven”, atp 2005, Issue 4
 J. Bausa, G. Dünnebier, “Durchgängiger Einsatz von Modellen in der Prozessführung”, Chemie Ingenieur Technik, 2005, Issue 77
 Rudolf Lobecke, Martin Zeller: “Betriebsführungssystem für eine Vielzweckanlage”, atp 2002, Issue 12
 “Plant Asset Management – betriebliche Wirklichkeit”; M. Gote, atp 2007, Issue 2
 “Von der Prozessführung zum Asset Management”, C. Maul, atp 2007, Issue 2
The author is head of the Process Management Technology Unit at Bayer Technology Services, Leverkusen.
This article is protected by copyright. You want to use it for your own purpose? Contact us via: support.vogel.de/ (ID: 220930 / Operation & Maintenance)