Process Improvement Tools Can Add Value in Turnarounds

(TMA International Headquarters)

The historic restructuring of the U.S. auto industry is reallocating human capital. As Adam Smith’s invisible hand works its magic, displaced engineers and executives will create tremendous value in other sectors of the economy. How will they do it? Here are five process improvement tools and methods that are scalable to a middle market turnaround engagement or private equity investment:

• Theory of Constraints (TOC) (systemic problem solving to avoid the local sub-optimization trap)
• Industrial engineering and lean thinking (basic blocking and tackling to reduce the seven wastes)
• Value Stream Mapping (VSM) (visualizing the seven wastes)
• Manufacturing/business process simulations (predicting future operational and financial results in multi-product /multi-process environments)
• Enterprise profit models (systems thinking game changers)

An understanding of when to apply these methods should be of interest to all turnaround management professionals and private equity executives. After all, everyone wins when innovative solutions are found to increase cash flow, reduce working capital requirements, and mitigate risk.

In the 10-machine manufacturing process illustrated in Figure 1, each machine had a team of manufacturing engineers managing its design and development. All 10 teams hit their performance target: 50 units per hour and 98 percent availability. A separate team of plant engineers developed the plant layout. Their objective was to develop a lean layout minimizing work-in-process inventory. By carefully arranging the 10 machines, they achieved perfect one-piece flow.

Before the new manufacturing process went into production, the teams briefed senior management on the status of the project. All involved had met or exceeded their objectives. Waste had been minimized. All the lean metrics looked great. Optimism for a successful launch was very high, and why not?

When the new manufacturing process was launched, production was only 42 units per hour and not the 49 units per hour that was expected. If all of those involved met or exceeded their objectives, what went wrong?

The interactions between the machines were not considered. When Machine 6 is down, Machines 1 through 5 are immediately blocked and Machines 7 through 10 are immediately starved. In isolation, each machine could produce 49 units per hour. In combination, they could not.

This manufacturing puzzle can be solved by inspection. But when one replaces the 10 machines with functional organizations, such as marketing, sales, scheduling, purchasing, manufacturing, distribution, and customer support, no longer are interactions (blocks and starves) quite so easily observed. In fact, the performance of the enterprise likely is being sub-optimized as the functions work hard to improve their local metrics.

And that’s the opportunity. First, one should think of business functions as part of a larger enterprise system and then apply these five continuous improvement methods to improve operational performance; increase earnings before interest, taxes, depreciation, and amortization (EBITDA); and create value.

Theory of Constraints

TOC is a proven process to solve the “10-machine” local optimization paradox. It views an organization as a chain of dependent activities or functions all working toward a goal. The constraint is the weakest link in the chain, one that most severely limits an organization’s ability to achieve higher performance relative to goal. In business, that goal is usually to make more money now and in the future.

Eliyahu Goldratt outlined the following process to improve organizational performance in his first book on TOC, The Goal :

• Step 0: Define the system. In this context, the system includes both the goal and the activities and functions that deliver the goal: Who and what contributes to EBITDA?
• Step 1: Identify the system’s constraint. Finding the constraint in a large, complex organization can be a challenge. A simple rule of thumb is: If a link in the chain is blocked, then the constraint is downstream. If a link is starved, then the constraint is upstream.
• Step 2: Decide how to exploit the constraint. How can one get the most out of the constraint? Approve overtime? Reduce setup times? Improve scheduling? Increase incoming inspection?
• Step 3: Subordinate everything else to the decisions made in Step 2. What can non-constraints do to ensure that the constraint is as productive as possible? Cross-train people? Improve quality? Take lunch and breaks at different times?
• Step 4: Elevate the system’s constraint. Add capacity if and only if the constraint’s performance truly has been maximized.
• Step 5: If a constraint is broken in Step 4, go back to Step 1. Repeat the process on the next constraint until the organization’s goal has been met. If the goal is open-ended, such as increasing EBITDA, then this process never ends.

Industrial Engineering, Lean Thinking

Once the constraint is identified, lean principles can be applied to increase throughput at the constraint and reduce operating costs at non-constraints. Lean thinking is a management philosophy focused on the reduction of the seven wastes:

1. Overproduction (making more than customer demand)
2. Motion (human or machine)
3. Waiting (human or machine)
4. Conveyance (movement from one location to another)
5. Over-processing (making features not valued by the customer)
6. Inventory (raw materials or finished goods)
7. Correction (scrap and rework)

The principles of lean manufacturing started with Henry Ford and were refined into what is known today as the Toyota Production System (TPS). When a delegation from Toyota visited the United States after World War II, they concluded that mass production was not suitable in post-war Japan. They were, however, inspired by a supermarket’s simple but elegant process for restocking shelves.

TPS is a process-driven, long-term philosophy of continuous improvement and waste reduction. It is based on a “pull” system to avoid overproduction and minimize inventories, a culture of getting quality right the first time, standardized work, and visual control.

Lean thinking is not just for automobiles. Many costs are assigned when a product is designed. Companies now apply lean thinking to reduce waste in product development, including standardized parts, modular components, and design review checklists.

Both manufacturers and service providers incur significant wastes as material moves into, through, and out of their operations. Applying lean thinking up front to packaging design, container density, and material flow can reduce material handling costs.

Lean principles are being applied in service industries such as healthcare. Some of the benefits include shorter patient waiting times, increased utilization of critical resources (beds, operating rooms, testing machines), reduced medical and billing errors, and more effective discharge planning.

Value Stream Mapping

VSM is a method to illustrate the seven wastes and to identify their sources. VSM provides a comprehensive view of all actions—value-added and non-value-added—required to bring a product or service to a customer. A VSM (Figure 2) includes product flows (across the bottom) as well as information flows (across the top). In addition to illustrating process logic, a VSM organizes key data for each process step, such as cycle times, changeover times, lot sizes, uptimes, scrap rates, inventory levels, inventory delays, transport times, and shipping frequencies.

A VSM is a good first step in thinking systemically. Taking a value stream perspective ensures working on the big picture and therefore avoids the “10-machine” local optimization paradox. VSMs are equally valid for manufacturing, service, and administrative processes. Once a current state VSM is completed, it provides managers and employees an effective tool to find constraints and discuss alternative actions to reduce waste.

Software programs have been developed to facilitate the development of value stream maps. While nice to have, they are not essential. VSMs on brown paper covered with sticky notes are just as valid.

Process Simulation

While a VSM is a good first step in thinking systemically, the method has one significant limitation: it provides a static snapshot, not a moving picture.

A simulation provides a dynamic view of the value stream as well as the ability to run “what-if” experiments to predict future operational and financial results. As such, simulation is a very powerful and versatile tool to maximize investment efficiency and mitigate risk in both manufacturing and business processes.

Discrete event simulations (Figure 3) are built by connecting modeling elements such as machines, conveyors, buffers, parts, and people in the process flow logic. Next the performance of each element is described with variables, such as cycle times, downtimes, changeover times, conveyor min/max/floats, buffer sizes, and shift hours. By happy coincidence, most of the data required to build a discrete event simulation already has been organized on the VSM.

Uncertainty in any performance variable can be captured by fitting a probability distribution around a mean value. By using a different random number stream for each probability distribution, the events in the model are independent of each other, just as in the real world. At the end of a run, the simulation software collates the results and generates reports. What-if experiments are easily performed by making changes to the input data set, rerunning the model, and then comparing the results.

Many manufacturing and business processes share resources in complex ways (Figure 4). In such cases, finite capacity simulations can create multi-product, multi-process production schedules for improved customer service, reduced inventories, and better utilization of resources. Finite capacity simulations are equally applicable to manufacturing and service organizations and can be interfaced with shop floor and human resource systems. Finite capacity simulations have similar dynamic what-if capabilities. What if a shipment of parts is running late? What if a machine is down for the day?

For more than two decades, discrete event simulations and finite capacity simulations have proven equally adept in finding constraints and testing strategies to break them through the Goldratt five-step process.

Enterprise Profit Models

At the top of the ladder are enterprise profit models (Figure 5). As their name suggests, enterprise profit models cross organizational chimneys, integrate manufacturing and business processes, and focus on a company’s financial objectives: profits, cash flow, return on net assets, etc.

Enterprise profit models incorporate both micro- and macroeconomic factors: price elasticities, fixed/variable cost structures, supply chain logistics, and capacities/staffing, as well as aggregate demand, tax policies, exchange rates, and tariffs. A modular model architecture is typically used. Enterprise profit models are too big in scope and in diversity for a single simulation methodology.

Typical problems addressed by enterprise profit models include order-to-cash, build-to-order, and product portfolio/plant footprint optimizations. Constraint-based continuous improvement methods still apply but on a much larger scale.

Enterprise models require the close collaboration of senior business leaders across functional chimneys. To optimize enterprise performance, non-constraints must subordinate themselves to the constraint. Not all functional chimneys can maximize or minimize all of their local performance metrics.

Case Studies

Once the constraint is identified through TOC, some combination of the remaining four tools can be used to improve operational and financial performance. Here are three representative case studies:

Furniture Factory. The upholstery area was identified as the constraint in a furniture factory. A team of industrial engineers developed a VSM. A rearrangement of the upholstery cells was proposed to reduce waiting losses. A discrete event simulation model was developed to verify the proposed changes. Next, a finite capacity scheduling model was developed that reduced setups while still meeting delivery lead times. Combined, the rearrangement of the upholstery area and resequencing of fabric types doubled throughput and reduced headcount by one operator.

Medical Center. A medical center needed to determine the minimum number of operating rooms to equip and staff. Patients arrived in the system each day and were classified as emergency or elective and as cardiology or non-cardiology. Emergency patients automatically were treated on the day they were generated, and elective patients were selected for treatment if there was time available in the operating room before the end of the shift. Cardiology operating rooms could service either cardiology or non-cardiology patients, although cardiology patients had priority.

A discrete event simulation model was developed to study operating room capacity alternatives. The model predicted overtime, operating room utilization, patient throughput, and patient waiting time. By optimizing the number of cardiology and general operating rooms, the medical center delayed buying costly equipment and hiring staff until the level of patient demand required it.

Durable Goods Manufacturer. A consumer durable goods manufacturer had a good problem on its hands: its new plant in Brazil had many more orders for a hot new product than it could make. An enterprise model of the South American order-to-cash process was developed. Cross-functional teams quickly discovered that minimum market allocations were limiting profits—a policy constraint (Figure 6).

A fact-based method was developed to estimate minimum dealer survival volumes across export markets. This freed up capacity that could be allocated more profitably based on such factors as margins, exchange rates, shipping costs, and tariffs. Next, targeted investments were identified in the plant to better align product mix with customer demand—a physical constraint. When the Brazilian real (R$) appreciated in relation to other South American currencies, exports were reduced in favor of the nation’s domestic market. Thanks in part to the order-to-cash enterprise profit model, quarterly profits increased by $126 million from the previous year.