Intelligent Manufacturing € July € 1995 € Vol. 1 € No. 7

Beyond Transactions:
What Operations-Oriented Software Can Contribute to the Bottom Line

By Julie Fraser

Between reengineering, downsizing and empowerment, manufacturing organizations are getting flatter and flatter. No longer are all decisions made by a few top executives. At more and more critical junctions, operations and line personnel determine what course the company takes.

The questions to ask yourself are:

• Have my systems changed to reflect the new organization?
• Do my operations people have decision support tools?
• Can they determine from those systems how to leverage resources to the maximum?

Line personnel must have a different type of tool than traditional computing applications, which, as the name "computing" suggests, concentrate on number-crunching. Notice how this legacy lives on: many MIS organizations report up to the CFO. Most traditional systems, even MRP, capture all the transactions, and provide good planning and data visibility for financial roll-ups.

Two issues lead us to the conclusion that we need more than traditional systems: 1) traditional systems are used by a certain group of managers and planners who work in offices and make decisions on a longer timeframe than operations people; and 2) capturing data and providing visibility only indirectly supports people making decisions; direct decision support would help more employees make fast decisions.

Operations peoples' primary job is to add value, not to analyze data; the tools must aid them to add value in the most rapid and effective way. Thus, the systems must be more responsive, and targeted at optimizing, quick what-ifs and tradeoff analysis, not just visibility to mounds of data.

What's interesting is that operational software, used well to support decision making, will have a much higher and faster return on investment than traditional, financially-oriented software. By looking at the differences in the two types of software, we'll see how it happens that operational systems contribute more to the bottom line than traditional systems can.


System Differentiators
Characteristics of systems to serve operations (or primarily non-financial) users are different than traditional systems (see Table I). Unfortunately, many MIS departments only have experience with the former. Operations personnel are familiar with control systems that fit the latter description. They may not realize, however, that there are now operational systems that work in plant-wide situations and in close conjunction with traditional systems. Examples include finite scheduling and synchronization software, asset care or maintenance management software, and product configurators.

Traditional systems are called management systems; they provide visibility into data about the operations. These new systems leverage the data traditional systems manage - about resources, suppliers and operations - to actually improve the effectiveness and productivity of the plant operations.

Let's take the example of manufacturing synchronization and scheduling software. The plan from the transaction system serves as input, including customer orders with their due dates, quantities and mix; supplier lead-times and qualifications, etc. The scheduling system includes a model built with actual data about production resources, their interdependencies, resource calendars and activities. The scheduling application uses algorithms on the data and model to deploy the resources in a way to balance customer on-time deliveries with plant resource utilization.

To talk about the differences another way, an MRP system will give you a comprehensive view of all of the data you enter into it. An operations system such as a scheduler will, based on the data you feed it, optimize the operation for higher performance and higher profits, giving back an answer that even the most expert human could not have created alone.


Bottom-Line Contributions
One of the business drivers for most manufacturers is cycle time reduction. The reasons come down to profitability and financial measures. It's well proven that reduced cycle time results in:

• quicker response to customer demand (agility),
• lower work-in-process (WIP) inventories,
• higher percentage of resources used on invoiceable production, which can be measured as sales dollars per resource,
• lower finished goods inventories, and
• higher return on assets.

Transaction systems can reduce the data manipulation time, but cannot otherwise affect the cycle time of the business process. Traditional MRPII and plant data collection systems can give a view into inventory levels, cycle time and problems, some time after the fact. In a simple plant, people may be able to use this information to make decisions. However, as customers drive us to produce a higher variety, while at the same time providing shorter delivery cycles, lower costs and higher quality, most plants offer so many variables that people are overwhelmed. A person cannot even consider all of the relevant data, and because of the timing, data is often so old that it's not relevant.

Operational systems take that data and add logic to show the best decisions. For example, manufacturing synchronization and scheduling products can actually reduce production cycle time with resource deployment optimization. This type of optimization software includes logic to ensure that resources are working on orders that are due soon and can be completed, based on other resources being available. As you can imagine, this minimizes cycle time.

To compare the approaches of traditional and optimizing software to a similar problem, take for example a plant that has one machine with a long setup time that is used near its capacity. When an important new order comes in, either type of system could run a new schedule. However, the results could have different financial impacts:

• A traditional transaction system simply distributes the load, and may show that the machine is over capacity. It thus recommends that you increase capacity through overtime, outsourcing, or purchasing another machine. In short, it recommends you spend more money. What you don't know is whether the materials, qualified personnel and tooling are available if you do decide to run overtime, for example. So you could be throwing good money after bad. Further, the system does not give any direction on when to insert this order. If a new setup is done resulting in other orders shipping late, you could lose business from several customers forever, just to get one order out.
  
• A finite scheduler with manufacturing synchronization capabilities would examine the schedule and try to readjust other work to get all orders out on time. It would try to group the new order in with others that require the same setup on the bottleneck machine. It would examine alternate routings, and ensure that the bottlenecked machine was fully utilized earlier, too. Usually, it would find a way to schedule the new order and keep the existing orders coming out on their due dates. If it did not, however, it would give a view of what the trade-offs are, what other orders would be late if this one is inserted (and how late so you can notify customers), and allow the human scheduler to determine what trade-offs to make.

Good operations systems allow manufacturers to set business rules and choose the balance they want between customer responsiveness and profitability. By viewing all of the assets, and globally optimizing for a mix of orders, these manufacturing synchronization solutions can leverage the investments every manufacturing company has already made: in their people, their equipment, their tooling, and in transaction systems to manage data.

Plant and line personnel have always made many of the decisions that determine the profitability of a manufacturing company; now, executive managers are openly encouraging that. Fortunately, there are a range of software tools that serve the needs of this new breed of decision-makers.

Traditional or Financial Systems	Operational Systems
Transaction-Oriented	Decision Support-Oriented
Planning-Oriented	Execution-Oriented
Infrequent Updates Needed	Frequent Updates Needed to Reflect Change
Aggregates, Averages, Standards	Actuals, Calculations, Results
Data Recording- and Visibility-Focused	Optimization-Focused

Julie Fraser is corporate strategic advisor for Berclain (Schaumburg, Ill.), a provider of software and methods for manufacturing synchronization & scheduling to discrete and batch process manufacturers. She can be reached at (508) 362-3480, or by e-mail: [email protected]