**Problem Definition**

In the last coordination meeting, the Management asked the quick budget comparison of the carbon steel (CS) pipeline 4” with the Polyethylene (PE) Pipeline 125mm with length 6 km. Technically, both pipeline already fulfill the management requirement. Because of we already have data from the previous project, author want try to calculate with Cost Estimating Relationship (CER) method.

**Development of Feasible Alternatives**

A parametric model is a useful tool for preparing early conceptual estimates when there is little technical data or engineering deliverables to provide a basis for using more detailed estimating methods. Parametric estimating is reliant on the collection and analysis of previous project cost data in order to develop the Cost Estimating Relationship (CER). A CER is a mathematical model that describes the cost of an engineering project as a function of one or more design variables. CERs are useful tools because they allow the estimator to develop a cost estimate quickly and easily.

There are four basic steps in developing a CER :

- Problem definition.
- Data collection and normalization
- CER equation development
- Model validation and documentation

Author will compare, which pipeline material has a better price to construct 6 km pipeline using Cost Estimating Relationship (CER) method, CS pipeline or PE pipeline.

From the 2015 project, we have data like table below (cost in Million USD) :

Table 1. Cost Data of CS Pipeline Project

Table 2. Cost Data of PE Pipeline Project

**Development of the Outcomes for Alternative**

The Indexes Method is one of the way to normalize above data into year 2017. Because of the limitation of index data (the author can’t get the 2017 data), the author use indexes from Chemical Engineering Plant Cost Index (CEPCI) as per below :

Figure 1. CEPCI Data of Feb 2016, Jan 2016, & Feb 2015

The author use ‘Pipe, valves & fittings’ category of CEPCI for both the. For year 2015, the CEPCI = 863.2, and for year 2017, because of the author don’t have the 2017 indexes data, the author use the February 2016 CEPCI = 791.2.

The factor = 791.2/863.2 = 0.92

The normalization of both data are per below :

Table 3. Normalized Data of CS Pipeline

Table 4. Normalized Data of PE Pipeline

After normalization, We develop CER equation using regression function in excel.

The result of CS Pipeline are :

Figure 2. Regression Result of CS Pipeline

The result of PE Pipeline are :

Figure 3. Regression Result of PE Pipeline

**Selection of the Acceptable Criteria**

The selection of the criteria is which pipeline material has a better price (lowest price) to construct 6 km pipeline.

**Analysis and Comparison of the Alternatives**

From the regression result in step 3, the CER equation for pipeline are :

- Carbon Steel 4″ Pipeline : cost = 6.01 + 104.47 x

Where x represents the length of pipe in km, and 0.18 ≤ x ≤ 7.40

- Polyethylene 125mm Pipeline : cost = 12.39 + 102.86 x

Where x represents the length of pipe in km, and 0.50 ≤ x ≤ 9.00

Using the CER equation cost for 6 km pipeline are :

**Selection of the Preferred Alternative**

Based on comparison table above, Author recommend the PE Pipeline for this project.

**Performance Monitoring and Post-Evaluation of Results**

Documenting the development of CER, including the related data is important for future use. Actual data from next or another project will become very useful for CER validation.

**Reference:**

- US Government, Department of Energy (DOE). (2011).
*Cost Estimating Guide*. Washington,D.C., Chapter 5, 19-21. Retrieved from : https://www.directives.doe.gov/directives/0413.3-EGuide-21/view - Sullivan, W.G., Wicks, E. M., Koelling, C. P. (2014).
*Engineering Economy*, Chapter 3, page 103 to 117. Pearson. Sixteenth Edition - CEPCI Indexes. Retrieved from https://www.researchgate.net/post/Where_can_I_get_2016_chemical_engineering_plant_cost_index_CEPCI
*Cost Estimating and Assessment Guide: Best Practices for Developing and Managing Capital Program Cost*, GAO-09-3SP. Washington, D.C.: March 2009, Chapter 11, 112-118. Retrieved from : http://www.gao.gov/new.items/d093sp.pdf

Dear Dr Paul,

Please accept my w6 blog posting.

Still struggle to revise w5 blog posting and w7 blog posting.

Best Regards,

Fakhri

AWESOME pak Fakhri!!! Really nice work on this one. AND there is good news. If you wish, you can take exactly the same case study but for your W7 blog, instead of limiting it to the PE and CE pipelines between 0.2 and 9 KM what I would like to challenge you to do is try using BEST FIT curves and EXTRAPOLATE your data out to say 15 or 20 KM?

To see what I am suggesting that you do, go see Ibu Irene’s W6 posting http://emeraldaace2017.com/2017/09/10/w6_ai_price-forecasts-for-offshore-regasification-facility-project/ and/or go see what Steve did with his data. https://js-pag-cert-2017.com/w14_sjp_forecasts-part-6/

Bottom line- you really have picked some truly outstanding case studies and now I want to challenge you to take them to the next level of sophistication.

BR,

Dr. PDG, Jakarta, Indonesia

PS Pak Fakhri, don’t just use linear regression….. Try to find the three “best fit” models which are usually linear, polynomial and exponential….. But you will have to experiment around to find out which give you the “best fit” understanding you are looking for an R^2 value of >0.90, with the highest value being your “most likely” outcome.

Then you can apply PERT to come up with a P75, P85 or P90 values for longer distances…

Going to be a VERY powerful and important tool that you end up developing based on your actual data.

Looking forward to seeing your W7 blog on this topic….

BR,

Dr. PDG, Jakarta