Jump to content

Photo - - - - -

Assumptions – Their Necessity And Their Pitfalls In Engineering Design

When I first started working in engineering design I came to know the real significance of the word "assume".
But can we avoid assumptions? The practical view point is that you cannot avoid assumptions during the initial phase of engineering design. Let us understand why it is necessary to assume during the initial stages of engineering design.

When engineering of any project is started one of the main tasks is to get the specifications / datasheets of long-lead items sent to the manufacturers as soon as possible. The long-lead items are the ones which require a long duration (typically 12-18 months) to engineer, manufacture and transport and without which the plant / unit being constructed cannot be commissioned or started up. Essentially it means that these items need to be given top priority during the engineering design phase by the engineering consultant / contractor. Any delays in procurement of the long-lead items could prove disastrous for the financial viability and profitability of the project and thus a delay or hitch in their procurement is a strict no-no.

Why assumptions are required? At the start of the project a lot of input data required to perform engineering calculations is not available. For example, if the layout of the proposed plant / unit an the various equipments in it is not frozen and still under debate and you need to prepare a datasheet of a pump which is a long-lead item, then hydraulic calculations (e.g. NPSHa, rated discharge pressure, differential head) need to be performed irrespective of the fact that no line routing (suction / discharge) data is available. A tentative location for the pump suction source, pump and the pump destination and pipe routing need to be assumed. The hydraulic calculations need to be done based on the above-mentioned assumptions.

What happens if the pump and its suction and discharge layout changes drastically from the assumed layout and pipe routing? Your hydraulic calculations become totally useless and you are back to square one.

Another example of an assumption would be the assumption of physical properties of a chemical for which information is available for a single set of conditions. Many chemicals are sold by trade names and many of the manufacturers provide data such as vapor pressure, specific gravity, viscosity for a single temperature value. If that chemical is being pumped at a different temperature then the one provided in the manufacturer's datasheet, then can you assume the properties for a different operating temperature? Any experienced engineer would tell you that this is being foolhardy.

With lack of data at the initial stage of the project, how does one proceed to do engineering? This is where the well established and bigger engineering consultants / contractors come into the picture. A lot of these reputed engineering consultants have a well established reference database of similar past projects and vendor data to kick-start the project on a satisfactory note. The assumptions in such cases are not just pure assumptions conjured out of thin air but are based on previous data of a similar nature and are much more reasonable then any random assumption. These engineering consultants also have direct contact with the equipment manufacturers whereby any unforeseen changes in engineering data at a later stage can be absorbed without much impact on the equipment delivery schedule.

One of the often neglected but very important aspects in engineering is maintaining a database of past projects and vendor database.

The reputed engineering consultants in the chemical process industry have developed their own internal engineering design standards for engineering based on accepted international standards and good engineering practices. This helps these companies to execute any project better and faster.

To conclude, assumptions are unavoidable during the initial phase of the project. However, the assumptions made need to be based on past experience and sound engineering logic and cannot be random assumptions. Unsound assumptions can have disastrous consequences on the project in terms of implementation, operation and safety of the plant / unit.

Hoping to have comments on this blog entry from the rich pool of talented chemical engineers who subscribe to the "Cheresources" forum.



G Vishwanathan
Oct 30 2010 07:27 AM
Another challenging "assumption" is with regard to Heatexchanger design pressures in the Pump Circuit.The pump curves would not be available in the initial stages of Datasheet preparation and many times young Process engineers will put "Hold" on the design pressures of Tube and Shell.This is not acceptable to the heatexchanger designer of mechanical discipline.
To overcome this problem,it is better to look for a similar pump both with respect to Head and Flowrates and use the pump curves to look at the Shutoff Pressure.This with a factor of safety should be able to solve the problem.
G Vishwanathan
Qalander (Chem)
Nov 12 2010 12:58 AM

Really Appreciate this alerting blog, Best of Luck

Best Regards

Nice introductory discussion.  There is further detail regarding the process of managing assumptions available on Wikipedia.  It is written from a business/project management point of view, but I feel the general principles are applicable to managing technical assumptions, too.


It is worth noting that the identification and listing of assumptions is only part of the task - they must then be quantified in terms of their criticality (impact potential), tested (validated) and tracked if not closed out.  This process is an integral part of effective technical risk management.

Latest Visitors

  • Photo
    Today, 06:06 AM
  • Photo
    Yesterday, 05:03 PM
  • Photo
    Yesterday, 07:49 AM
  • Photo
    02 Mar 2024 - 15:25
  • Photo
    02 Mar 2024 - 11:07
  • Photo
    01 Mar 2024 - 22:33
  • Photo
    01 Mar 2024 - 14:24
  • Photo
    01 Mar 2024 - 09:16
  • Photo
    01 Mar 2024 - 07:22