OR Tools that Help Level the Playing Field (Stage 2: Re-prioritize your Goals)
You may first want to test your assumptions about the market niche you're after. There are at least two Operations Research techniques that assist you in this effort. The first, sometimes referred to as the analytic hierarchy process (AHP) , allows the management team to develop a comprehensive ranking of several alternatives in each of a number of decision criteria, in order to determine the most likely outcome of any management strategy. For instance, a lumber company might be evaluating whether or not to expand into the dimension business, and in so doing be looking at three or four different ways to go about it. Several criteria might be chosen, such as initial capital requirements of each option; availability of experienced personnel; impact on cash flow; and number of potential new customers for each option. The management team then determines a weighting of the options with respect to each criteria through pairwise comparison relative to a qualitative scale, such as extremely preferred, strongly preferred, equally preferred, etc. After several iterations of this process, a priority matrix is produced which gives the management team a fairly objective view of the options they are considering.
Another more technical way to go about this process involves a technique referred to as goal programming . Goal programming works well where the issues under consideration are strictly quantifiable, but it does not deal with subjectivity as cleanly as the AHP.
Another issue that may lead many companies into double jeopardy is that of inaccurate product costing. Many companies' market strategies are based on mistaken assumptions on the costs of their individual products. Therefore, the operational goals they are trying to meet may be inappropriate, and contributing to their acceptance of customer orders that other, larger companies avoid.
The actual cost of products is a fairly simple calculation when companies are small and product lines limited. However, many companies find that with the success and growth of their business, the actual costs of their products become blurred as new raw materials are found, employees are added, and improved manufacturing techniques are developed. Most companies eventually get into traditional cost accounting techniques, which allocates fixed and variable costs across products as a function of the volume of each product.
However, in the complex manufacturing of many wood products, and especially in the business of secondary wood products, traditional cost allocations can be very misleading and cause companies to mis-price their products due to incorrect assumptions. If you find your profits per product line less than they used to be relative to the market, then cost accounting may be contributing to the problem. A new technique developed in the 1990's, called activity-based costing (ABC) , has proven superior to traditional cost accounting in those businesses where the manufactured products consume different levels of resources, or use more activity, to express it another way. Many companies have completely re-thought their product line offerings in light of the findings of an ABC implementation.
Once an operational strategy is decided upon, optimal implementation becomes the issue. Many companies talk about "optimizing" their operations, but very few are truly optimizing based on quantitative goals and constraints. Author and business consultant Gene Bryan, in his book The Best Possible Sawmill , in commenting on the complexity of the sawmilling business decision-making process, notes that the following are "the right things" to be considered:
· Maximizing profit instead of production
· Buying, selling, and trading the right logs
· Choosing the right log-breakdown patterns
· Running work centers the right number of shifts and hours
· Investing available capital where it will do the most good
· Pursuing the right markets and customers
· Making the right amount of each product
· Selling the right mix of products.
In order to properly, objectively, and quantifiably consider these things, Dr. Bryan suggests, requires at least the following data:
· Log availability
· Production rates and yields
· Manufacturing and other variable costs
· Sales prices and market opportunities
· Inventory levels
· Customer commitments.
Once these data are gathered, an OR technique called linear programming (LP) can be used to properly model the system, and the opportunity for every business is to develop, maintain, and use the implemented linear program on a regular basis to contribute to the regular assessment of market opportunities and to truly "optimize" its operations and production schedules to meet those opportunities.
Another aspect of success in delivering the right products to market at the right time is anticipating product demand. There are two general approaches to this: forecasting demand in advance, and building Just-in-Time (JIT) response systems to actual orders.
Forecasting market demand is usually done intuitively or historically through sales projections, but there lies a great opportunity to apply statistical techniques such as various forecasting models. Many investment companies use these tools to anticipate specific movements of the stock market; we can utilize the same tools to anticipate timber or lumber prices, cabinet or furniture sales, or even strategic investments by competitors.
The JIT approach at first appears to eliminate the need for forecasting, as a company implementing a JIT strategy and the associated lean manufacturing techniques focuses on building a rapid-response system to customer orders and eliminating all waste (including in-process inventories) associated with non-order activities. Indeed, lean manufacturing techniques are an essential component of remaining competitive on the world market; the challenge to us in the wood products industry is to understand which of the many components of lean give us the most leverage, and which, if any, do not apply so neatly to our problems of naturally-induced raw material variability (wood!).
In fact, I view statistical forecasting tools and JIT strategies as complementary techniques that work best when both are used. It's too long a topic for this TechNote, but rest assured that any manufacturing strategy done with a certain knowledge of future conditions can deliver better results than one that waits on events to unfold.
Finally, as you come to understand how your goals need to be re-prioritized, you must address the issue of employee development. Many companies enter into these projects thinking that reduction of employee headcount is one of the inherent advantages. In fact, I have found the opposite to be true…that companies who get really serious in the use of these tools come to realize the benefit of re-training many of their employees to use these new tools. Employee jobs tend to change, but the total number of jobs doesn't usually decrease. In fact, one of the signs that an OR implementation is truly successful is that new jobs are created where the tools have identified and optimized new opportunities. After all, growth is a sign of a healthy, competitive organization.
I'll conclude this series in the next TechNote by discussing tools that are essential in the successful re-allocation of your resources: process metrics, process sampling, project management techniques, and employee involvement. See you again in a couple of weeks.
Chuck Ray, Ph.D.
Penn State Wood Products Operations Specialist