When it comes to creating an optimal manufacturing process that limits variation and conserves energy or resources, or a developing a new formula that is most likely to meet customer expectations, design of experiments (DOE) is an indispensable tool.
In the chemical industry, manufacturers use DOE to develop processes that help reduce waste, improve their environmental footprint and identify the best substitutions for ingredients affected by market scarcity (or changing environmental or healthy/safety regulations). Many of these same motivations for using DOE apply to other industries, such as the food and beverage industry – or even electronics manufacturing.
DOE can help manufacturers improve their processes or find ingredient substitutions that are more likely to be successful using fewer experiments or test runs.
During a pandemic or disruption in normal distribution chains, DOE can be invaluable. DOE can help manufacturers more effectively create a new recipe to replace an ingredient such as sugar or olive oil if it becomes scarce or hard to obtain due to shortages or distribution challenges.
Benefits of DOE in Manufacturing Settings
For manufacturing industries, coming up with the most efficient process parameters that reduce waste and production costs, while delivering the most robust result can’t be left to speculation. Engineers can make estimates about what the optimal settings should be, but without data analytics, your processes won’t be optimized. You need very precise indications of the correct process parameters that reflect the specificities of the manufacturing equipment.
You could guess the best settings, assess the results, and then try to further improve the settings by modifying one factor at a time. But this could be a lengthy process, and you still might not find the optimal settings.
Even if you have the optimal settings for current conditions, to ensure that the plant remains at the leading edge, new technologies or ingredients will need to be adopted — introducing unknown situations and new problems. Ultimately, you need a comprehensive model that helps you understand, very precisely, how the system works. You obviously cannot get that by adjusting one factor at a time and experimenting in a disorganized way.
You need a tool that lets you build models in a very practical, cost-effective and flexible way. Design of Experiments (DOE) is the ideal tool for that.
Using DOE in the Food and Beverage Industry
Similarly, in the food and beverage industry, being able to create a new recipe or establish an effective production process without having to run hundreds of experiments can save time and money. In addition, DOE can help overcome the biases or limitations that exist when relying on taste panels for product development decisions.
The failure rate for new products in the food and beverage industry is high. Only about 10 percent of all new products on market are successful.
DOE can help to increase your chances of success. You can’t always rely on what people say they like so using data to inform your decisions makes sense.
Some problems that exist with relying on taste panels when making decision about product recipes include:
- Your panel may not be representative of the consumers you want to sell to
- Your trained panel may see thing differently than your consumers
- The taste profiles between different countries could be different. (So, if you’re focused on three markets, such as US, Turkey, Sweden, you might need three different recipes. DOE in combination with multivariate data analysis (MOCA) will help find the intersection between all the taste profiles.)
By taking all the factors into account and creating a way to analyze various elements of taste and their past effects on market success, DOE can help create a product that is more likely to succeed in the market. You can include these factors in your design of experiments, and then see how changes in the recipe are directly related to high panel likes.
DOE can help you define the ideal recipe that:
- Fulfills consumer requirements
- Can be produced at reasonable cost
- Is not sensitive to changes in raw materials
How Does DOE Help?
DOE removes a lot of manual work involved in testing various options. It helps speed up the process. Using a DOE software tool (such as MODDE ) can also make it easier, because users can focus on the problem they’re trying to solve, rather than having to learn a method or software. Software tools like MODDE with wizards guide the user, even if not experts in data analytics, to make finding the right set of experiments that define the outcome easier.
For example, beginning with idea development, DOE will tell you how likely you are to meet the requirements that are defined by your panel and at what cost you can do that. By taking those responses into consideration and creating the right tests, you can define the minimum viable product.
DOE helps you to:
• Minimize the number of experiments you have to do to find the ideal recipe
• Create a robust process (one that holds up to changes in environment, humidity, etc.)
• Adapt a recipe for changes in ingredients or packaging needs (availability, environment, regulations, consumer trends)
Minimize Number of Experiments
DOE suggests which variables to test and defines the number of runs (often fewer) needed to confirm it. DOE results in a model or an outline for which factors need to be tested to ensure the most robust formula, process or timeline.
The typical alternative to using DOE to determine which process or formula recipes are possibilities is the classic “COST” approach. The COST approach, which stands for “Change One Separate factor at a Time” is a logical way to approach an experimentation, but it requires a lot of time and effort. DOE helps shorten that process and uses advanced data analytics to help inform your decisions about which factors need to be tested.
One of the weaknesses of the COST approach is that it can’t define the number of runs that will be needed to find the maximum. You can only test two factors at a time and can’t define how many runs you’ll need to be sure you’ve covered all the options. You also can’t know what happens if you have a different starting point. With DOE, you can address these issues. It provides a direction to follow and allows you to incorporate many difference factors into the equation while using fewer tests.
DOE requires fewer experimental runs to achieve the same - or even better - results than the COST approach does.
Define Formula or Process Robustness
DOE can also help define how robust a recipe or formula will be. How does it react to a variation in raw materials? If a harvest one year is not the same as the year before, or an ingredient is not available, DOE can help define a recipe or formula that is less impacted by that ingredient change. This reduces the chances that a slight change in manufacturing conditions, such as a change in temperature, storage or humidity, would affect production or quality, helping you maintain consistency.
Find Substitute Ingredients
In the chemical industry, moving toward more green chemicals or environmentally safe options can be driver in the need to change a process or formula. You might need to replace petrochemicals with biochemicals or change from first to second generation feedstock, such as sawdust, wood chips or waste streams from food that might otherwise be thrown away. DOE helps you find the right strategy for adjusting your formula with fewer test runs.
For food and beverage manufacturers, DOE helps identify the process or ingredient that keeps the product the most consistent with customer expectations. For example, if you vary the sugar content of a specific recipe (or replace it with a sugar substitute) that can not only affect the taste of the product but also a number of physical properties as well. It may look different, be darker or lighter in color, may become more brittle or have a shorter shelf life.
DOE provides a systematic way to predict outcomes of your changes with fewer test runs. It can also take into account customer preferences in ways that relying on tasting panels can’t.
Application Areas for DOE
Data analytics, including DOE, provides valuable input for many stages of product development.
- In the strategy/idea development stage, DOE is important for concept engineering.
- In the product design and process development stage, DOE supports product design, consumer testing, product optimization and process development.
- As part of commercialization, DOE and data analytics are important for market testing, HACCP processes, self-life analysis and process engineering.
- During launch and afterward, it supports continuous process improvement and production monitoring and improvement.
DOE can support your process at all levels from idea development to product design to manufacturing.
DOE can support your product development at all stages of product development from idea development to product design to commercialization.
Want to Know More?
Watch one of these recorded webinars to learn more about using data analytics, DOE and process optimization in process industries. Including:
- Design of Experiments (DOE) for the Beginner
- Improved Profitability and Operations in Pulp and Paper Plants
- Optimizing the Operation of Boilers
- Using a Data Driven Approach to Design Next Generation Manufacturing Processes
- Using Six Sigma for Process Improvement