Chemical Engineering MagazineChementator Briefs
H2SO4 Catalyst BASF SE (Ludwigshafen, Germany; has introduced a…
12. Alfa_Laval_Valve_MatrixFocus on Valves
This butterfly valve has an inflatable seat for tight closure…

Comment PDF

Cost Engineering: Integrating Technology and Economics

By Thane R. Brown |

The Process Synthesis Model presented here is a methodology that integrates technology and economics, thereby improving process design and reducing costs

Broadly, the chemical process industries (CPI) are comprised of companies in the chemical, petroleum refining, pharmaceutical, household and personal-care products, paper products, and the food-and-beverage and other sectors. Each sector is continually faced with product and process obsolescence, with cost inflation, and with the growth of health, safety and environmental (HSE) regulations. To neutralize these forces and remain competitive, CPI companies must continually develop or synthesize new products and upgraded processes.

This article is about process synthesis, and it presents a methodology that integrates technology and economics, thereby improving design quality and reducing costs. Spending for process synthesis occurs in three areas: research and development (R&D), capital investment and production. R&D expenses include the staffing, laboratory, pilot plant, and related efforts involved in doing synthesis work. Capital investments include the costs to design, install and startup a “synthesized” process. Production costs consist of the raw materials, packaging materials and manufacturing costs that are needed to produce a product made by a “synthesized” process.

While there is no concrete estimate of synthesis costs, the author’s experience suggests it is in the range of 10–30% of R&D, 10–25% of capital and 10–30% of production cost spending *. Table 1 shows what CPI companies in the 2015 Fortune 500 spent in these three areas. Using those data, U.S. synthesis spending in 2015 is estimated at between $100 billion and $250 billion/yr. Worldwide spending is of course much greater. For perspective, worldwide chemical sales [1] in 2015 were three times U.S. sales. Ratioing up U.S. synthesis costs by three approximates worldwide synthesis costs of between $300 billion and $750 billion.



The methodology in this article, dubbed the Process Synthesis Model, has the potential to reduce synthesis-related spending by between 5 and 10% per year while also improving design quality. The cost savings will vary from company to company, dependent upon a company’s type of business and its focus on product and process upgrades. For the sectors in Table 1, a 5–10% reduction in synthesis costs would decrease spending by $10 to 25 billion/yr. Globally, it’s between $30 and 75 billion. Use of the Process Synthesis Model has the potential to yield major profit improvement and cost reduction while also improving design quality.


BOX: Process synthesis objectives for Product X

The following is a summary of the key objectives and other information used for the examples (adapted from Figures 7.1 and 7.4 of Ref. 6).

Business objectives
Business plan. There is a market opportunity for a low-cost olive oil substitute made from soybean oil, Product X. Develop a process for Product X.
Projected volume. At this stage, potential volume is uncertain.  Estimates range from 200 million to 700 million lb/yr.  We will have to do further consumer testing to more accurately estimate volume.  

Timing. Complete the development work so we can begin the national introduction of Product X by late 2020. Product for consumer testing will be needed. See the consumer testing schedule.

Economic factors. We wish to sell Product X for no more than a 25% premium to our existing products. Develop the process accordingly.

Technical objectives
Business need. Develop the hydrogenation process including catalyst selection for the olive oil substitute, Product X.

Schedule. Pilot plant construction must be completed within six months. This should permit process development to be complete by January 2018, enabling a start of national production in late 2020. Sample product for consumer testing must be available by July 7, 2018.

Economic factors. Production costs should not exceed $0.35/lb. 

Technical factors.

  • See the Product Research Report for details about customer needs, product attributes and specifications. In it, you will find details about the bench scale testing that proved we can hydrogenate refined soybean oil to yield 98% conversion of C18:3 chains to C18:2 chains, 85% conversion of C18:2 to C18:1, and negligible (<0.05%) conversion to C18:0 chains. Trans-isomers formation was also less than 0.05%. The reaction used the new C-760 catalyst. Design the process to yield these conversions.
  • The catalyst must be removed from the oil after hydrogenation so that the oil is clear. Preliminarily, this will mean no more than 30 ppb left in the oil. To meet the production cost target, the catalyst will have to be used multiple times.
  • The projected capacity (200–700-million lb/yr) would indicate a continuous process. However, you should work with the plant design engineers to determine whether a batch or continuous process will be best from an economic standpoint. This will probably involve deciding how many process locations are best.
  • Plant operation is expected to be 24 hours per day, 7 days per week and 50 weeks per year.
  • Health, safety and environmental. All regulations and company policy will be followed. Since the hydrogenation catalyst contains heavy metals, your work must consider how to properly dispose of or reclaim the catalyst.


The methodology

Process synthesis is defined as the creation of a process that will make a product meeting customer-based quality specifications, HSE regulations and business and economic requirements.


The foundations of synthesis

From the definition come the three foundations of synthesis: product quality, HSE requirements, and business needs and economics.

Product quality. Consumer needs define product attributes, which in turn define product quality requirements. In their book, “Chemical Product Design” [2], Cussler and Moggridge discuss how to define customer needs and translate them into quality specifications.

Product quality specifications are different for different kinds of products. For example, consider the following:

  • Chemical attributes include items such as product purity, types and levels of impurities, color, odor, density, viscosity and particle size distribution
  • Food specifications cover items such as flavor, mouth feel, texture, color, viscosity, appearance and particle size distribution
  • Pharmaceutical specifications consist of things like drug purity, types and levels of impurities, taste, color, viscosity, rate of adsorption by the body

HSE requirements. HSE involves what occurs both inside and outside a plant. Inside the plant, HSE deals with protecting people, equipment and the facility. Outside, it has to do with the community and the environment. At a minimum, companies should strictly comply with all laws and regulations and do that in a cost-effective way.

To ensure the protection of the community, employees, equipment, facilities and the environment, work during synthesis has to consider the following:

  • Health hazards: allergens, carcinogens, mutagens, toxic substances
  • Personnel safety hazards: fire, explosion, flammable and explosive materials (which include some dusts), overpressure and others
  • Equipment and facility hazards: fire, explosion, flammable and explosive materials, overpressure and others
  • Environmental discharges (toxic and non-toxic, as well as planned and accidental), to sewers, rivers and lakes, the atmosphere, landfills and so on

Business needs and economics. Most companies exist for one reason — making money. To make money, they have to develop and operate processes that reliably deliver customer-driven product quality at a competitive price while economically complying with HSE regulations. In the more profitable companies, engineers begin integrating technology and economics during synthesis and continue that focus throughout the life of a project. Integration brings a strong economic focus to the technical work, which results in a climate that supports identifying and analyzing design options. “Experience has shown that when leaders maintain such [an economic] focus throughout a project, project teams can reduce capital costs by 5–10% …” [3]

Of interest is a study done by the Construction Industry Institute (CII). While its study dealt with the later stages in a project (FEL 2 ** and beyond), its conclusions also apply to synthesis activities. CII concluded better design quality, reduced costs and shorter schedules are statistically correlated with the following:

  • Setting clear, measurable project goals before work begins
  • Creating and studying a comprehensive list of options early in the project and continuing that throughout the project. Doing this produces a powerful economic focus within the project
  • Developing a project plan centered on early, firm scope decisions. (These early scope decisions are the ones made during synthesis and are the ones upon which a project is based)
  • Staffing a project with the right number of appropriately experienced engineers

Per the CII report conclusions, the best-organized projects had capital costs 4% lower than expected and 20% lower than the more poorly organized projects [3]. These data support the earlier estimate of cost savings and profit improvement from using the Process Synthesis Model.

When thinking about technical and economic integration, consider the following:

  • Finding the “best” technical solution, rather than just a workable solution, requires creating and studying a list of options more comprehensive and thorough than is typical [3]. Better decisions and designs emerge when the synthesis team broadens the number of options it considers and when the options include more than the team’s favorite solutions [4]
  •  The cost structure of a product and its process is largely set in the early stages of a project. Thus, it is crucial to combine economic and technical decision making from the beginning of synthesis [5]
  • In the author’s experience, most engineering problems have more than one technically workable solution. When that is the case, a good tactic is to use economics to decide which of the workable options is best [6]

cost engineering

Figure 1. This schematic shows an overview of the Process Synthesis Model (Adapted from Figure I-1, Ref. 6)

Process synthesis— a model

The Process Synthesis Model (PSM), illustrated in Figure 1, integrates technical and business and economic decision making. It is designed around the following three principles:

  • Setting objectives, both business and technical, before work begins
  • Identifying more design options and ideas than is typical
  • Using economics to select the best of several technically acceptable design options

Defining objectives. Just as one decides where he or she wants to go before starting out on a roadtrip, the synthesis leader sets objectives for his or her project before the actual work begins. Objectives marshal the energy of those doing the work and focus it toward the same endpoint. When work begins without clear objectives, the following two things happen:

  • Objectives are sorted out by trial-and-error. The trial and error approach produces changing focuses, inefficiencies and wasted effort and time
  • Unneeded work is done, work that will be shelved or thrown away

Good objectives, which need to be realistic and measurable, are created using input from the key stakeholders. First, the synthesis leader works with business management to learn what business results they want. Since business managers are busy and are pulled in many directions at the same time, they often have difficulty fully engaging in the business needs discussion. A face-to-face interview with the manager is a practical way to get around this issue. Phone interviews also work, but are not as effective.

Second, the synthesis leader and his or her technical boss translate the business objectives into technical objectives. These objectives, stated in technical terms, are what guide the day-to-day work of synthesis. To complete objective setting, the synthesis leader writes the objectives down and sees that each stakeholder gets a copy. The box above (Process synthesis objectives for Product X) is an example of the two types of objectives for a new edible oil process.

The synthesis team should regularly assess progress toward its objectives. If a goal is not on track, the team develops corrective plans. Lastly, it reports status to the stakeholders.

For more details about setting objectives or about the interviewing method, see Chapter 7 of Ref. 6.

Creating options. Creating options is the heart of the PSM. In this part of synthesis, a tone of exploration and of possibilities is needed to develop a thorough and complete option list. Because criticism, idea evaluation and option selection get in the way of idea exploration, it is best to hold these activities until after list completion.

Building an option list can be difficult. The following three tools make the task easier and will result in a higher-quality list.

  • The technical function flowsheet (TFF) is a flowsheet creation method that fosters divergence while defining the flowsheet. This method sets the stage for more complete option lists
  • Probing questions are a series of questions intended to uncover more options by deepening one’s understanding of the process
  • Unit operation (UO) guides are lists of common UOs arranged by types of technical functions, such as “react a gas with a liquid” or “separate a solid from a liquid.” They greatly simplify identifying UO options

The technical function flow sheet. The TFF is a block flow diagram that shows all the steps in the process and all the major flow streams. It is a bit different form of block flow diagram in that each process step or block is described in terms of technical functions, not unit operations or equipment. Technical functions specify the transformation taking place within a block, for example: heat oil, hydrogenate oil, or separate catalyst from oil. Note these are verb-noun pairs. Corresponding unit operations would be a U-tube heat exchanger, a jacketed reactor, and a pressure filter, all of which are adjective-noun pairs.

Smith suggests beginning flowsheet creation with the reactor, moving to the separation and recycle systems, and lastly defining the HSE processes [7]. When working on the reactor and separator, one also defines the feed preparation for those blocks. If a process does not include a reactor, begin with the central operation in the process. If there is a reactor upstream or downstream of the process, it is wise to look at how the upstream reactor affects the process or how the process affects the downstream reactor.

Often significant time passes between the synthesis of a process and its installation in a plant. This time lapse makes it important not to overspecify the process during synthesis. The following guideline takes this into account and enhances the technical and economic potential of the process: “Specify as little as possible while ensuring the process will operate as intended” [8]. This allows those working well after the completion of synthesis to bring current technology and economics to bear on the selection of unit operations and equipment.

There are times, however, when the synthesis engineer must specify UOs or equipment, particularly for the reactor and separator and functions to ensure the process will operate as required. Again, it is best to leave the added definition as open as possible. For example, one might spell out two or three options, rather than selecting a single UO. Of course many functions need no added specification. For example, heating and cooling functions are typically simple applications of heat exchange technology.


Examples: Product X

The examples in the rest of the article will be based upon Product X, which was introduced in the box above (Process synthesis objectives for Product X). These examples illustrate creation of a TFF, development of an option list, and option analysis.

Product X is a low-cost olive-oil substitute, made from soybean oil. The company developing it has extensive experience with edible oils processing — refining, hydrogenation, fractional crystallization and deodorization. The price of soybean oil averages ~$0.40/lb, and olive oil ~$1.95/lb.

The company’s R&D department has made small quantities of Product X in the laboratory by selectively hydrogenating soybean oil. After hydrogenation, the oil has a composition very similar to olive oil. The R&D engineers catalyzed the reaction with <0.5 wt.% of a proprietary, finely divided, nickel/zirconium/tantalum catalyst (C-760). The reaction was carried out at 20 psig and used ~10% excess H 2.The gas-to-liquid (G/L) ratios are ~9 at the beginning of the reaction and ~2 at the end. The reaction is irreversible and exothermic.

Soybean oil is a glycerol molecule with three hydrocarbon chains attached. The chains have 16 or 18 carbons atoms and are a mixture of saturated carbon bonds — designated as C16:0 and C18:0 — and unsaturated bonds (one, two, or three) designated as C18:1, C18:2, C18:3. The weight percent compositions of soybean oil, olive oil, and Product X are shown in Table 2.


Example 1: The TFF

Create the TFF for Product X. The first step is to define the reaction and reaction feed steps. The synthesis engineer initially defines the reaction function as: “Hydrogenate soybean oil using C-760 catalyst so that 98% of the C18:3 is converted to C18:2, 85% of the C18:2 is converted to C18:1, and negligible C18:0 and trans-isomers are formed.”

From talking with those who did the initial bench work on the reaction, the engineer verifies the reaction selectivity will be hard to achieve. Thus, the engineer decides to run bench and pilot-plant tests to better define the conditions that will deliver the required selectivity. While running these tests, he or she concludes there are only two reactor options that can reliably produce Product X, either a continuously stirred tank reactor (CSTR) or a batch reactor. Since both are technically feasible, both are included in the reactor block. The revised reactor function description is now: “Hydrogenate soybean oil using C-760 catalyst so that 98% of the C18:3 is converted to C18:2, 85% of the C18:2 is converted to C18:1, and negligible C18:0 and trans-isomers are formed. Use a batch or CSTR reactor.”

Figure 2.  Three versions of the technical function flowsheets (TFFs) for the oil-hydrogenation example are shown here: preliminary issues 1 and 2 (top and middle), and the final issue (bottom). The figures are adaptions of Figure 8.2 and 8.6 of Ref. 6

Figure 2. Three versions of the technical function flowsheets (TFFs) for the oil-hydrogenation example are shown here: preliminary issues 1 and 2 (top and middle), and the final issue (bottom). The figures are adaptions of Figure 8.2 and 8.6 of Ref. 6

Next, the engineer finalizes the reactor feed function as, “Heat oil to reaction temperature.” Figure 2A shows the flowsheet after the definition of the reactor and feed portions of the process.

The next phase of synthesis involves separation. There are two separations, removal of excess H 2 and separation of catalyst from the oil. The two technical function descriptions are: “Separate H 2 from the oil,” and “Remove catalyst from the oil so it is clear.” Figure 2B shows this version of the TFF.

Further work yields the completed TFF, Figure 2C. The changes versus Figure 2B are as follows:

  • An expanded function definition for catalyst removal, which now includes catalyst reuse. Because of the very small particle size of the catalyst, oil-catalyst separation is difficult. After running several tests, the engineer decides to specify either a pressure filter or a tubular-bowl centrifuge to ensure proper separation. The function description becomes: “Remove catalyst from the oil for reuse and so oil is clear. Use a pressure filter or a tubular-bowl centrifuge”
  • Recycle and purge streams were added, namely the recycle and purge of catalyst for reclaiming and the purge of H2 Environmental items were added, namely purging H2 to the atmosphere and the reclaiming of spent catalyst
  • A function for reactor cooling.
  • A function for slurrying catalyst in feed oil
  • A function for cooling the oil to storage temperature

The TFF is not a stand-alone document. Also needed are material and energy balances. When synthesizing a block flow diagram, Murphy proposes using a Degree of Freedom analysis when doing the material balance [ 9]. In addition, each flowsheet function requires added definition. This is done via a document called the Technical Function Definition (TFD). It details the following:

  • The purpose of each function
  • Important variables and operating conditions
  • The basis for the variables and for any unit operation that was specified (see Table 3 for the TFD that accompanies the final issue of the flowsheet).
  • The TFD is the forerunner of the process definition, which is a detailed prose description of the process


Option creation tools. As already mentioned, generating a high quality, comprehensive option list can be difficult. Most of the difficulties are linked to the engineers’ mindset and to organizational pressures. Generally, engineers have good problem-solving skills, which enable them to quickly solve technical questions; but in terms of option generation, this gets in the way. Once there is a solution, there is no need to look at other options. Engineers also tend to have a production focus, which creates internal pressures to “get on with the job” and not take time to explore multiple options.

The main organizational issues are schedule pressures and standardized designs. Schedule pressures push the engineer to quickly make design decisions, not spend time looking at options. I have found, however, that more fully examining options often simplifies execution and may shorten the schedule. Standardization conflicts with the philosophy of considering options. If thorough option analysis was done prior to setting the standard and freezing the design, standardization could yield big economic benefits. On the other hand, standards do periodically need updating, and the standardization policy makes this hard to do.

There are forces that hinder option list development. To counter these, this section of the article presents tools that make the option creation task easier and faster. Two are useful in all situations, brainstorming, and a variation, the “6-3-5” method [10]. Both are small group tools that foster creativity by allowing group members to build upon each other’s ideas. The Process Synthesis Model also uses two other tools, Probing Questions and Unit Operations Guides. Both lead to more thorough lists.

Probing Questions. These questions will help the synthesis team better understand their process and uncover more options. The questions are arranged as follows: general, feed preparation, reactor, separations, recycle/purge, HSE and references.


  • Could the product be purchased? If so, why not buy it rather than make it?
  • Should the process be batch or continuous?
  • Does the process require materials of construction more expensive than carbon steel? If so, how might the corrosive streams be eliminated or their concentrations be changed to reduce corrosion problems and to lower costs?
  • Does the proposed process require operation outside the ranges of 0 and 150 psig or 100 and 650°F? If so, how might the process be changed to permit operation within these ranges, lowering costs?
  • Should there be surge between unit operations (UOs)? How much?
  • What are the technical and economic tradeoffs among the main operations on the flow sheet — reactor, recycle and purge streams, separation systems and heat recovery system? In general, the reactor design determines how the separator will be designed and what recycle or purge streams will be required. These, in turn, affect the heat-recovery plan and systems. All have an effect upon the utility needs [11]

Consider each technical function (TF), as follows:

  • What final product attributes does this TF determine?
  • What attributes of the input stream(s) will this TF change?
  • How can one manipulate this TF to control the output stream or final product attributes?
  • How do the upstream TFs and processes affect this TF?
  • How does this TF affect the downstream TFs and processes?
  • How could one change the attributes of the input stream(s) to have a drastic impact on the performance or cost of this TF and those downstream?
  • Can the product specs be relaxed and would that make the process less expensive to build and operate?
  • Does this TF have a significant impact on capital or production costs, either by itself or by its effect on downstream operations? If so, are there other options available [for example, changes to the input stream(s), changes to the output stream(s) specification(s) or changes to the final product stream(s) specification(s)]?

Feed preparation:

  • What different grades and sources of raw materials are available? What effects do these have on reaction rates, on reaction yields, on byproduct formation, on human or environmental safety, and on process operation and costs?
  • Does the feed stream contain materials that should be removed before being processed? (For example, those that poison or foul the catalyst or materials that form difficult-to-separate or hazardous products/byproducts)
  • Does the feed stream contain a large amount of impurities?
  • Does the proposed process use feed streams or create intermediates or other streams that are health or safety hazards or that require control or treatment for environmental reasons? If so, how might the process be changed to eliminate or minimize these issues and their associated costs?


  • Is there more than one reaction path? Which has the best kinetics, yield and so on? Which uses the least expensive raw materials? Which produces the least amount of byproducts? Which has fewer or no HSE issues?
  • What reaction conditions maximize reactor yield, selectivity and first-pass conversion? Consider the reaction phase, temperature, pressure, concentration of reactants, degree of agitation, reaction rate and so on.
  • Does the reaction need to be catalyzed? If so, will the catalyst be solid or homogeneous? Will the catalyst have to be removed from the product stream?
  • If there is a reversible byproduct, should it be removed or recycled?
  • Which reactant should be in excess? Consider that effect on recycle/purge. If there is an excess of one or more reactants, should they be purged, recycled, or recovered?
  • Should there be more than one reactor system? If so, does there need to be a separation system in between the two systems?
  • Is the reaction exothermic? Could it become a runaway reaction? If so, what can be done to eliminate or control this hazard?
  • How should the reactor be operated, isothermally or adiabatically? If adiabatically, is a runaway reaction possible?
  • How should the reactor temperature be controlled — no control, with a recirculating fluid; with an excess reactant; with an inert diluent; or with cold or hot shots?
  • Will the reaction temperature be near or within flammability limits? If so, what safety precautions are needed? Could another reaction path be used to eliminate this hazard?


  • What is the purpose of the separation: purification of a component, removal of undesirable components, recovery of a component for sale, further processing, or recycle?
  • What purity is desired for the different streams leaving the separator system? How does purity affect the choice of separator and separation sequence?
  • Is the separator feed is a heterogeneous mixture (solid-liquid, gas-liquid, other) or a homogeneous fluid? This affects the type of separator and the sequencing.
  • What are the possible separation sequences for the removal of components from a process stream? Consider the phase of the feed stream — single phase or multiphase
  • Do the differences in boiling points, melting points, solubility in various solvents, adsorbent properties lend themselves to enabling separation?
  • If solvents are used, should they be miscible or immiscible in the feed?
  • If adding something to the feed, a solvent or an adsorbent, are there HSE issues with the material?
  • Are there reactive or corrosive components in the separator feed? If so, consider removing them quickly.


  • Should streams to be removed in the separator system be recycled or purged?
  • Should the recycle stream(s) be purified before re-entering the process?
  • Should the purge stream(s) be treated or reclaimed?
  • What is the optimum amount of recycle or purge?


  • Are there materials used or made in the process that are hazardous for HSE reasons or that require environmental treatment? If so, can they be eliminated, be used in reduced quantities, or be replaced by less or non-hazardous materials? If these materials are reaction products, can reactor conditions, recycle amounts, or the catalyst be changed to reduce the amount generated?
  • Can hazardous materials, essential to the process:
  • Should HSE considerations be a critical siting factor? Should the process be located:

One will find more detail on HSE in Ref. 7 and Refs. 12–15.

References.The question list was developed using Refs. 6, 7, 9, 12, 13 and 16.


Unit Operation Guides

When it is time to select a UO for a technical function, the need to develop options again surfaces. The UO Guides allow one to pull together a list in a few minutes. Had these UO Guides been available when I was designing plants, I would have made a number of different UO selections.

There are eight UO Guides: Blending and Mixing, Drying, Heat Transfer (including Evaporation), Mass Transfer (including Crystallization), Material Transport, Mechanical Separation, Reactions and Size Modification. They list the more common UOs and are arranged by the phases of the materials being processed. Additionally, each Guide includes a reference list.

To go about creating a list of UO options, one uses the Guide corresponding to the technical function involved. For example if the technical function is to “heat oil,” one would use the Heat Transfer Guide; or if the function is to “hydrogenate oil,” one would use the Reaction Guide going to the column that correlates with the phases being processed. That column lists the UO possibilities for the engineer’s situation. An example illustrates.


Example 2. Create a list of unit operation options for the technical function in Figure 2B, “Remove Catalyst from the Oil.” Since catalyst removal is a mechanical separation, use The Mechanical Separation Guide (Table 4). Being separated are oil (a liquid) and a solid catalyst, so one would use the Liquid-Solid column, which lists the following options for the separation function:

  • Clarifier/thickener (three types)
  • Screens (four types)
  • Floatation systems
  • Expression presses
  • Hydroclones
  • Sedimenting centrifuges (three types)
  • Filtering centrifuges (three types)
  • Pressure filters (four types)
  • Vacuum filters (three types)

In total, there are 23 options to consider when one includes all the sub-types. This is a sizable list. Some might call it overwhelming. However, I have found most options will be eliminated during a preliminary technical evaluation. This will be shown in Example 4 (online version, below).


Analyzing and selecting options

The analysis of options is a process of elimination. The synthesis engineer first does a technical analysis of each option, assessing which meets all of the quality, HSE and business requirements. Any option not meeting all of the requirements is eliminated.

The challenge is to correctly evaluate the options without doing a detailed analysis of each. So, one begins with a high-level, cursory review, which will eliminate many to most of the options. As the evaluation continues, it becomes more detailed, until only a few options are left. Quite often, in the latter stages of evaluation, the engineer will need more data and will have to do laboratory, bench-scale or pilot-plant testing. Only then can he or she determine whether an option is technically feasible. The statistical design-of-experiments methodology enables the efficient design and analysis of experiments [17].

At the end of the technical evaluation, there is usually more than one technically acceptable option. When that is the case, the engineer uses economics to pick the best of the technically acceptable options. Most often having only one option is optimal, but there are times when it will be appropriate to carry more than one forward, at least for a while.

Economic analysis requires estimating the capital and production costs for each option. Using this information, the engineer calculates the net present value (NPV) or the annual cost (AC) for each and picks the option having the highest NPV or AC (Note: NPV and AC measure the same thing; they just express it differently. Think of AC as the annuitized form of NPV). When only two options are being compared, one can also use return of investment (ROI), choosing the option with the highest ROI. However, when comparing three or more options ROI is cumbersome and should not be used. A few guidelines will be helpful when doing an analysis.

  • Only compare costs that are different among the options. For example, if raw material costs are different, include them in the analysis. If energy costs are the same, exclude them
  • For the discount rate, use the company’s hurdle rate
  • To select the economic life for the comparison, use obsolescence or the company’s financial guidelines. Use the same life for each option
  • Designate the year in which revenues or cost savings begin as year one
  • Use after tax cash flows for the NPV and AC calculations

Chapters 5 and 10 of Ref. 6 fully explain the economic comparison method.

Example 3: Technical analysis. You have identified two broad options for the reaction function. Do a high-level technical screening of both. The options are as follows:

  1. A highly selective hydrogenation reaction that will produce the desired product composition. It requires a very expensive, proprietary catalyst, C-760. The catalyst needs further development to optimize the ratios of nickel, zirconium and tantalum and the particle size. The later affects the removability of the catalyst from the oil.
  2. A less selective hydrogenation that uses a much cheaper commercially available catalyst, C-42.

Evaluation of Option 1. After reviewing the bench-scale data from the catalyst and reaction experiments and talking with the engineers who did the work, the engineer verifies that the hydrogenation will reliably produce the desired composition for Product X. The synthesis engineer also estimates the optimization of the ratio of the metals and of the particle size have a 95+% probability of success and deems this option feasible.

Evaluation of Option 2. By talking with company engineers who have fractional crystallization experience, the synthesis engineer quickly confirms that the C18:0 can be crystallized while leaving the C18:1, C18:2, and C18:3 oils in solution. He or she also concludes the C16:0 will crystallize with the C18:0. Thus, during crystal separation both will be removed from the liquid oil. As well, he or she finds separating C16:0 from C18:0 from is not practical since their melting points are too close to each other. After the crystal removal step, the engineer calculates the oil would have the following composition:

  • C16:0 <1%
  • C18:0 <1%
  • C18:1 88%
  • C18:2 10%
  • C18:3 <0.2%

Since the Option 2 product contains no C16:0, it does not meet the desired composition for Product X. (Olive oil contains about 13% C16:0.) Although the oil without the C16:0 would be deemed healthy, it could not be marketed as an olive oil substitute. Therefore, it is eliminated.

Conclusion: Only Option 1 is technically feasible. Option 2 will be dropped from further consideration.

Examples 4 and 5 can be found in the online version of this article (see below references).



Use of the Process Synthesis Model will result in better designs and lower R&D, capital, and operating costs. The model has the following three phases:

  • Setting business and technical objectives before synthesis begins.
  • Creating and investigating a more thorough and complete set of design options than is typical. This increases the odds of finding more elegant solutions to any design problem. The Process Synthesis Model uses three tools to assist with option creation: the Technical Function Flowsheet,
  • Probing Questions and Unit Operations Guides
  • Assessing options. One first evaluates options technically and eliminates those that do not meet all of the quality, HSE, and business requirements. If there is more than one technically acceptable option, one uses economics to determine which is the best option



3Thane Brown (Email: worked for more than 36 years for Procter & Gamble in a variety of engineering and manufacturing roles, primarily in the food-and-beverage business and in health, safety and environmental engineering. In his last position there, Brown was director of North American engineering. After retiring, he taught engineering economics at the University of Cincinnati, and plant design at the University of Dayton. Brown is presently a member of the Chemical Engineering Advisory Committees at the University of Dayton, at Miami University (Oxford, Ohio), at the University of Louisville and at the University of Cincinnati. He also works as a SCORE counselor, providing free assistance to small businesses in the Cincinnati area. Brown authored the book “Engineering Economics and Economic Design for Process Engineers” [6], as well as a number of articles on engineering economics, batch pressure filtration and heat transfer. He is a registered professional engineer in Ohio (inactive), and holds a B.S.Ch.E. from Oregon State University.



1. Tullo, A.H., Global Top 50 Chemical Companies, Chem. & Eng. News, July 27, 2015, pp. 14–17.

2. Cussler, E. and Moggridge, G.D., “Chemical Product Design,” Cambridge University Press, U.K., 2001.

3. Brown, T.R. and Singh, S., Project Optimization through Engineering, Chem. Eng. July 2014, p. 51.

4. Heath, C. and Heath, D., “Decisive,” Crown Publishing Group, New York, 2013.

5. Brown, T.R., Capital and Production Costs: Improving the Bottom Line, Chem. Eng. January 2010, p. 26.

6. Brown, T.R., “Engineering Economics and Economic Design for Process Engineers,” CRC Press, 2007, p. 201.

7. Smith, R., “Chemical Process Design and Integration,” John Wiley and Sons, 2005, p. 9.

8. Ref. 6, pp. 151–152.

9. Murphy, R.M., “Introduction to Chemical Processes: Principles, Analysis, Synthesis,” McGraw Hill, New York, 2005, pp. 121–137.

10. Ulman, D.G., “The Mechanical Design Process,” McGraw Hill, New York, 1997, pp. 147–148.

11. Smith, R., “Chemical Process Design,” McGraw Hill, New York, 1995, pp. 3–8.

12. Douglas, J.M., “Conceptual Design of Chemical Processes,” McGraw-Hill, 1988.

13. Turton, R., others, “Analysis, Synthesis, and Design of Chemical Processes,” 4th Edition, Prentice Hall PTR, 2012.

14. Peters, M.S., Timmerhaus, K.D., and West, R.E., “Plant Design and Economics for Chemical Engineers,” McGraw Hill, New York, 2003.

15. Perry, R.H. and Green, D.W., 7th Edition, “Perry’s Chemical Engineering Handbook,” McGraw Hill, New York, 1997.

16. Seider, W.D., others, “Product and Process Design Principles Synthesis, Analysis, and Evaluation,” 3rd Edition, John Wiley & Sons, 2009.

17. Kleppmann, W. Optimizing Products and Process Efficiently, Chem. Eng., November 2014.


Online-only content

Example 4: Technical analysis. Example 2 identified twenty three UO options for removing catalyst from oil. Complete a preliminary technical evaluation of the options.

To start the analysis, the synthesis engineer lists the performance needs of the catalyst removal function.

  • Process objectives: clarifying and solids recovery
  • Capacity: Using the rough capacity estimate from the business objectives in the box above (Process synthesis objectives for Product X), the engineer calculates the separation system will have to operate between 50 and 200 gal/min
  • Solids settling rate: <0.1 cm/s (determined from a bench-scale test)
  • Solids in the feed: <0.5 wt.% (as stated in the technical objectives, box above)
  • Cake growth rate (if filtered): The engineer calculates that at a capacity of 200 million lb/yr, the filter cake growth rate will be in the slow range (cm/h)
  • Separator performance: The supernate must have good clarity. Since the catalyst will be reslurried in oil, the filter cake only needs to have fair cake dryness
  • Batch versus continuous solids discharge. After considering this question, the engineer decides either batch or continuous discharge would be acceptable and removes this option from the list

To enable a comparison of the options, the engineer next constructs a table (Table 5) listing the performance characteristics of the different options. Note the lightly shaded columns correspond to the performance needs of this separation.


The engineer next assesses each of the options versus the characteristics in Table 5 and eliminates the options not technically feasible.

Eliminated: (In Figure 6, the factors causing elimination are shaded in red.)

  • Clarifier/thickener: the solids settling rate is lower than the minimum for clarifiers and the cake dryness is not good enough
  • Screens: the solids settling rate is too low and the supernate clarity is not good enough
  • Flotation: the additives needed to float the catalyst will contaminate the oil and catalyst
  • Hydroclones: the solids settling rate is too low and the supernate clarity is not good enough
  • Scroll centrifuge: the supernate clarity is not good enough
  • Disc centrifuge: cake dryness is not good enough
  • Peeler centrifuge: the peeler is intended for feed streams having a much higher solids concentration, the cake growth rate is too slow, and the supernate clarity is not good enough
  • Pusher centrifuge: the solids settling rate is too low, the pusher is intended for feed streams having a much higher solids concentration, and the cake growth rate is too slow
  • Worm-screen/scroll: the scroll is intended for higher throughputs, the solids settling rate is too low, the scroll is intended for feed streams having a much higher solids concentration, and the cake growth rate is too slow
  • Cartridge filter: this filter is not designed to recover solids
  • Rotary drum, horizontal belt, and rotary disc vacuum filters: the solids concentration in the feed is too low and the cake is not dry enough

Technically acceptable:

  • Tubular-bowl centrifuge: However, it is only a marginal choice because the centrifuge is intended for a feed rate of ~1–50 gal/min, and the needed process capacity is 50–200 gal/min. Ultimately, this is an economics question as capacity can be dealt with by using multiple units
  • Plate-and-frame filter
  • Vertical-element pressure filter
  • Horizontal-element pressure filter

From this high-level analysis, the engineer concludes there are four technically feasible units—the tubular-bowl centrifuge and the three pressure filters: plate and frame, vertical element, and horizontal element. For a preliminary design, any would be a good choice and could be selected. For the final selection, additional technical or economic data (or both) will be needed.

Example 5: Economic analysis. After further analysis, the synthesis engineer excludes the tubular-bowl centrifuge from the list of acceptable options for catalyst removal. This leaves three options remaining: a plate-and-frame filter, a vertical-element pressure filter and a horizontal element pressure filter. Select the best of these options using economics.

First, the engineer decides how the filters will be operated.

  • The plate and frame filter will operate at 120°F because it has to be manually cleaned
  • Since both the vertical-element and horizontal-element filters are totally enclosed and self-cleaning, they can be operated them at the discharge temperature of the reactor, 350°F. This greatly reduces filter area because the viscosity of 350°F oil is only about 15% of that for 120°F oil, which reduces the resistance to filtration. On a dollar per square-foot basis, these filters are about three times more expensive than the plate-and-frame unit.

Next, he or she sizes the filters and estimates the capital investment and operating costs for each option. These are shown in Table 6.


With this data in hand, the engineer is able to select the most economic option by inspection, rather than by having to calculate NPVs or ACs. As both the plate-and-frame and the horizontal-element units require more capital and have higher operating costs than does the vertical-element filter, the engineer selects the vertical-element unit as the best, the most economic of the three options.

Related Content
Optimizing the Design-to-Cost Cycle
By intensifying cooperation between process designers and cost engineers in the conceptual stage of plant projects, better-informed decisions can be…
Competitive Pricing of Process Plants
Accurate pricing of CPI plants involves integrating technical design with economic evaluation and accounting for many types of risk. Knowledge…

Chemical Engineering publishes FREE eletters that bring our original content to our readers in an easily accessible email format about once a week.
Subscribe Now
Trinseo Digitizes Control System Migration Projects to Achieve Fast ROI
Purdue University Saves $400,000 Annually with Local Vacuum Networks
Bag filter Housings/Vessels
Innovative Backwashable Media Filter
Automated Vertical Tower Filter Press

View More

Live chat by BoldChat