Asset Utilization and Production Loss Accounting has become a core manufacturing improvement driver throughout Hercules. This presentation describes the keys to success in implementation and achieving results: defining the approach to Asset Utilization, gaining management support, and describing issues that need to be considered when implementing Asset Utilization and associated improvement processes.
Overview of Asset Utilization at Hercules
Asset Utilization, a measure of production performance, has become a key manufacturing improvement tool in Hercules. It was adopted as the primary measure for the Corporate Reliability Team, in December of 1997, after learning of it in a benchmarking session with Rohm & Haas. Today, Hercules’ system is actively used at 29 manufacturing sites worldwide. It provides data for eliminating production losses at each site, thus increasing capacity with negligible capital investment. The measurement system is also an effective reporting tool and is used corporately to define manufacturing improvement objectives.
Asset Utilization is defined as Actual Production divided by Ideal Production, for any given period of time. The difference between the two constitutes production losses, which incidentally are inherent in any manufacturing operation. Hercules accounts for Asset Utilization in hours, which negates the impact of product mix. An ideal rate is defined for each product. Production is converted to “Ideal Hours” by dividing the production quantity by the ideal rate. Asset Utilization is calculated by dividing the “Ideal Hours” by the actual hours for the time period. Hercules considers it important to measure all production losses. This is done by identifying the “real” ideal rate for each product, and by accounting for every hour in the year. Losses associated with planned downtime are accounted for by capturing the data in an appropriate loss category, so we recognize the downtime as a loss that was taken in a planned fashion. Many gains can come from measuring and managing the “acceptable” losses: the planned downtime and the small losses that are not recognized as significant.
Operational Asset Utilization (OAU) is identical to Asset Utilization, except that it excludes Demand related losses. Demand related losses are either plant shutdowns or production rate limits that are planned for the purpose of inventory control. OAU is a better measure of plant performance, and its increase can be directly tied to an increase in plant capacity. The difference between Asset Utilization and Operational Asset Utilization represents available additional capacity for meeting increases in market demand.
The objective of increasing plant capacity is to realize additional profits through the sale of additional production. This is by far the largest direct financial benefit of improving Operational Asset Utilization, as Gross Profit increases by sales price minus variable cost (mainly materials and part of the energy cost). Hercules selected six manufacturing processes for extra attention in 2000. OAU increases at these six sites allowed actual production averaging 13% above 1999 capacity. In overall results, all production processes in one division increased OAU an average of 4% for the year.
There are many other benefits of improving OAU, even in a plant that is not operating at capacity. As production (and maintenance) problems are eliminated, plant costs decrease. There are fewer abnormal situations that can lead to safety or environmental incidents. The plant gains the ability to respond to unexpected sales orders, increasing confidence in production promises and affording the opportunity to gain market share. Plant morale increases as operators and mechanics recognize real progress in eliminating recurring problems.
Asset Utilization Database
Hercules developed a database to capture production and loss data for Asset Utilization. Use of the database prompts the operator to account for production losses. The loss information is categorized at several levels of detail, and over time affords Pareto style reports identifying the largest cause of losses. The database also allows losses to be tied to a piece of equipment, the cause to be identified, and written comments to be recorded. See figures 1 and 2 for example reports on Asset Utilization losses.
The most effective tool in gaining management support is to point out the incremental gross profit associated with a potential increase in production for a process that is (forecast to be) at capacity. I used a 5% capacity increase, to evaluate the potential gross profit increase for various plant sites. Capacity increases of 10 to 15% are commonly possible through Asset Utilization improvement, with very little capital required.
At Hercules, we were fortunate that the decision to implement Asset Utilization was made by the Vice President of Manufacturing. Quarterly meetings with his staff were essential to raise the visibility of the effort. In those meetings we reported progress and discussed the path forward. Site specific progress reports by plant managers proved to be a very effective motivator for other groups.
The plant manager’s support is needed when implementing at a manufacturing site. I found that even very well intentioned plant people focus mainly on those items about which the plant manager is asking questions. I also look for the plant to devote a full week to the task of implementing our database and data collection processes. This has proven much more effective than an approach where training is provided but the plant tries to implement in a less focused fashion. Performing a system level “Failure Modes and Effects Analysis” is a useful tool for defining meaningful loss categories and for gaining understanding and consistency in loss accounting.
I consider it essential to provide the plants with the ability to customize the database setup to their preferences. Hercules’ database allows the user to define date format (American / Military / European) and production measurement units. We standardize on the top level of loss categories, but the plants can setup unique sub and detail categories. Our database can also be used for either batch or continuous processes, just by tailoring the end time of each record to either the end of the shift or the end of the batch. The begin-time is always the end-time for the previous record. Each plant can also decide whether to enter production and loss information every batch, shift, day, or other interval. The flexibility afforded by these setup and use considerations has enabled the same database to be used effectively by almost any process.
Proper setting of ideal rates is another important consideration. Production personnel frequently consider nameplate or design capacity to be the goal. But these capacities include assumptions for downtime and less than optimal rates. Our practice has been to allow production to set the ideal rates (best historical batch time or shift rate for each product), and then to increase them when some production records exceed the ideal rates. It is essential that production comes to view the ideal rates as a measuring stick for identifying all losses, and not as capacity that management will demand in the future. It is also essential for management to understand that lower Asset Utilization figures mean that a plant has set the standards high and is using the system as an improvement tool.
Asset Utilization is a measure of production losses. It is not a measure of downtime for each piece of equipment. At Hercules, we account for the production losses in hours, using the following equation:
Asset Utilization Loss Hours may not be equal to the downtime for a particular piece of equipment. Sometimes surge capacity (or parallel equipment) allows the process bottleneck to operate during part of the equipment downtime, or at a reduced rate. In the case of batch processes, the equipment may not be needed during part of the batch process. Therefore, the operator will need direction on properly accounting for production losses. When the operator enters a record into Hercules’ Asset Utilization database, the database calculates total loss hours. The operator then can review the problems of the shift in light of the total loss hours, when accounting for the losses during the shift.
For meaningful data, records should be entered every shift by an operator or supervisor. If data is entered before a shift leaves, the person entering the data has first hand knowledge of the events during the shift. This bridges the communication gap between shift and daylight. However, training and follow-up are required to ensure that each shift accounts for losses in the same manner, for data consistency purposes. Production entry by a day supervisor does have the advantage of making it easier to be consistent in how losses are categorized. Most Hercules plants start out using a day supervisor to enter Asset Utilization records. The supervisor gains experience in loss accounting, before delegating the duty to the shifts. While some plants choose to keep data entry at the daylight supervisor level, other plants are very successful using shift supervisors or operators for that duty.
There are several pitfalls related to accounting for “No Demand” losses (downtime or reduced production rates, taken to prevent excessive inventory when sales are below plant capacity). For “No Demand” rate limits, we consider the difference between production at the ideal rate and production at the “No Demand” rate as a “No Demand” loss, which does not reduce the Operating Asset Utilization (OAU). However, when plants have production problems they are inclined to speed up above their rate limit to make up for any problems encountered, which makes it look like there were no other losses on the shift. This is cheating and not allowed! Seriously, we ask the plants to limit their production to that rate, or if necessary to meet production requirements to remove the “No Demand” rate limit.
Another pitfall occurs when downtime is scheduled due to “No Demand”. Production schedules necessary repairs to take place during this scheduled downtime, because that is the right thing to do. They do not want this repair work to count as an “Availability” (Maintenance related) loss, because they were scheduled down anyway. But it does make sense to back the downtime for necessary repairs out of the “No Demand” category and account for them as “Availability”. This ensures a proper accounting of plant capacity and available production. Our goal is to increase OAU, then push to sell the additional capacity. However, we don’t want to sell out our ability to perform necessary maintenance!
In some plants that operate with only two or three shifts, there is a tendency to enter the loss hours that the operator is aware of, then enter the remaining losses as No Demand. This limits the usefulness of the data, because the operator is never prompted to look for losses they didn’t notice. Frequently, the “normal” losses are very significant and can be reduced. It is very important to accurately account for the No Demand downtime, to calculate the remaining production loss hours, then to look for the cause. This, in effect, is a benchmarking of every record against the best possible production, which highlights where there are opportunities for improvement. Benchmarking against yourself: what a concept!
Gathering data is an important endeavor, for benchmarking and measuring progress. Just paying attention to data entry, problems encountered, and posting results, spontaneously leads to improvement. There are also some powerful continuous improvement processes that should be implemented. One is what I will call a behavior based improvement process. Improvements come from focusing attention on the item to be improved. Central to this process is holding daily meetings to review losses, where day to day issues are resolved. The daily meetings should be centered on the operator and mechanic, who are encouraged to deal directly with any issues within their ability. But the big issues that require more resources must be resolved, not only for the direct benefit but also for morale considerations. These big issues can be addressed in a weekly meeting of the resource managers, where they prioritize, assign resources, and monitor progress. The plant manager should actively monitor progress, ask questions, and remove barriers.
Another important improvement process involves a tactical approach to loss elimination. Operations, Maintenance, and Engineering should meet for a full day every quarter to analyze loss information, and develop action plans to address the largest opportunities. Task teams are sponsored to validate the problems and implement solutions. This group should continue to meet every two weeks to monitor progress and results.
There are many important factors for driving improvement through Asset Utilization. Success requires management support, commitment of resources, useful tools, and focused improvement processes. The benefits are well worth the effort and cost, as success provides substantial financial and work environment benefits.