The article is a paper evidently presented at the World Academy of Science, Engineering and Technology 65 2010, entitled A Simulation Model for Bid Price Decision Making by R.Sammoura, who is with the Industrial Engineering and Management Department, Beirut Arab University, Lebanon. It offers a mathematical model to enable bidders to more carefully hone their bid prices to achieve success without leaving too much on the table.
My extracts, edited and arranged to suit myself, are as follows, but you should read the source paper for better accuracy.
Abstract—In Lebanon, public construction projects are awarded
to the contractor submitting the lowest bid price based on a
competitive bidding process. The contractor has to make a strategic
decision in choosing the appropriate bid price that will offer a
satisfactory profit with a greater probability to win. A simulation
model for bid price decision making based on the lowest bid price
evaluation is developed. The model, built using Crystal Ball decision engineering
software, considers two main factors affecting the
bidding process: the number of qualified bidders and the size of the
project.
The behavior of contractors as a group (market conditions, number and identity
of competitors), individual contractor behavior (contractor
size, work and tenders in hand, availability of staff), and
behavior toward the characteristics of the contract (type and
size of construction work, bid related factors) are the main
factors influencing the contractor’s bidding behavior.
Since it is not usually an easy job to describe the bidding
process by a realistic mathematical model interrelating all the
above influencing factors, a simulation model for bid price
decision making based on the evaluation of the lowest bid
price at a pre-contract stage is developed. The model
considers two main factors influencing bidding behavior: the
project size expressed by the average bid price and the level of
competition presented by the number of qualified participating
bidders. However, in order to reduce extraneous factors that
may distort the study results, all the selected projects
constituting the data sample are of the same type (in the field
of road construction and rehabilitation projects), awarded
according to the same Lebanese formal bidding procedures,
and executed in the same Lebanese market conditions.
The data for the study were collected from the archived
records of Council for Development and Construction (CDR).
All projects selected for inclusion in the
study were in the field of road construction and rehabilitation
in Lebanon. They were publicly bid under a relatively uniform
and formal bidding procedure according to the Lebanese
tendering law. The data collected from a sample of forty-one
awarded projects focused on the value of the lowest bid price
for each awarded contract. It also included the number of
qualified bidders participating in the bid process and their
corresponding bidding prices covering a time period for the
years (1996-2006). Among these forty one awarded projects,
twenty three are completely executed, nine are still in
progress, and nine are not executed. These forty one awarded
projects comprised 275 bidding attempts.
The paper goes on through a series of analysis including stochastic and regression analysis that I don't understand but was regularly exposed to in stock market technical analysis.
Essentially, to my unscientific mind, he looked at the proven standardized data that the government had available from past solicitations and came up with a trend following algorithm of the kind that tries to give lie to the common phrase, "past performance is no prediction of the future". I am reminded of the many financial wizardry schemes concocted by rocket scientists in the financial industry whose blind faith in such schemes brought the world to its financial knees over the course of this last decade.
So, while I appreciate the work in the model, for my money, it would be of little utility in terms of trying to price a contract.
But, it occurred to me, what a wonderful tool the data collection he describes would be to help government reduce its costs of acquisitions. What he did was aggregate and publicize certain bid abstract sheets. Over the course of time, this data would fairly begin to suggest, broadly, trends in project costs, and at the recent margin, a fair indication of expected costs on current projects.
If the government provided this data on all contracting, that would give bidders sharper pencils since they will have a more accurate picture of market pricing. Think of the stock market. You know the likely price of a share of, say, Google, because the stock market makes available to you all recent sale prices. Where pricing data is available, market transactions are more efficient (as a general rule).
If, say, the government has been buying plumbing equipment, getting the same bids from the same players, would a new market entrant know whether to go to the time and trouble to bid? If the new entrant knows what the government has been paying, it probably would attract him, especially if his pricing can beat the old regulars. And the more widespread the information is cast, the more likely others will come along to join in, bringing down government costs.
In the same way I think the government ought to cull, aggregate and make public recorded sales of real estate so that market participants are more equally placed to trade, it occurs to me that public dissemination of historical and recent bid pricing, available from bid abstracts, could help reduce the price government pays for its goods and services and, as this paper discusses, construction.
One of the core principles of government contracting is generating competition. Nothing generates competition like transparent pricing mechanisms.
Just saying.
No comments:
Post a Comment