SlideShare a Scribd company logo
1 of 26
Download to read offline
Fair ranking in competitive bidding procurement 
A case analysis 
Michael Soltys 
KES 2014 
18th International Conference on Knowledge-Based 
and Intelligent Information & Engineering Systems 
IS13: Modelling with Qualitative and Quantitative Pairwise Comparisons 
Pomorski Park Naukowo-Techniczny (PPNT) Gdynia 
September 15, 2014 
Fair ranking - Soltys KES'14 PPNT Gdynia Title - 1/26
Fair and transparent procurement procedures are a cornerstone of 
a well functioning free-market economy. 
Competitive bid procurement involves suppliers of services 
competing for a contract. 
The traditional method involves the Owner issuing an Invitation, 
directed to a group of suppliers. 
Bidders who wish to participate submit binding oers to do the 
required service by a
xed rate and time and in accordance with 
the Owner's requirements. 
The Owner then reviews and evaluates all on-time bids and 
chooses one Bidder to do the work. 
The Owner and that Bidder then sign a contract to do the work. 
Fair ranking - Soltys KES'14 PPNT Gdynia Introduction - 2/26
General rules of bidding 
For the Owner, the implied obligations or terms of the Bid 
Contract will be: 
1. A duty of full disclosure 
2. A duty of fairness and good faith 
For the Bidder the implied term of the Bid Contract will be: 
1. Honor the rules of the Invitation. 
Fair ranking - Soltys KES'14 PPNT Gdynia Introduction - 3/26
If any of these implied obligations are broken, then the wronged 
party has grounds for a lawsuit for breach of contract. 
Note that in the case that we examine in this paper we make no 
claims whatsoever of a breach of contract. 
Rather, we claim that the rules put in place to rank the bids did 
not adequately select the best bid. 
Fair ranking - Soltys KES'14 PPNT Gdynia Introduction - 4/26
Reilly vs Contemporary 
In our case the Owner is the Ministry of Community Safety and 
Correctional Services, and we consider two Bidders: Reilly Security, 
and the second Bidder was Contemporary Security. 
While Reilly is an Ontario based company, Contemporary Security 
is a Vancouver-based subsidiary of an American company. 
All companies (a.k.a., proponents) were bidding for a contract to 
provide security during the Pan American games to be held in 
Toronto in 2015. 
Since the Owner is the Ministry, the party concerned regarding the 
$-value of the bid are the tax payers of Ontario; indeed, the rules of 
the bid should satisfy the tax payers that they are getting the most 
competitive bid that can ful
ll the obligations of the contract. 
Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 5/26
The Ministry conducted the evaluation of proposals in the 
following stages: 
Stage I consists of a review to determine which proposals 
comply with all of the mandatory requirements. 
There were only two proposals left at the end of this 
stage: Reilly and Contemporary. 
Stage II consists of a scoring by the Ministry of each quali
ed 
proposal on the basis of the rated criteria. Only 
proponents meeting a minimum score of 65% will 
proceed to Stage III. Stage II will be worth 35% of 
the proponent's overall score. 
Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 6/26
Stage III consists of a scoring by the Ministry of each quali
ed 
proponent meeting the minimum score in Stage II on 
the basis of a presentation. Stage III will be worth 
25% of the proponent's overall score. 
Stage IV consists of the sealed pricing envelope provided by 
each proponent meeting the minimum score in 
Stage II and completing Stage III will then be opened 
and Stage IV will consist of a scoring of the pricing 
submitted. Stage IV will be worth 40% of the 
proponent's overall score. 
Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 7/26
At the conclusion of Stage IV, the scores from Stage II, Stage III, 
and Stage IV will be added, and the highest scoring proponent will 
proceed to Stage V| where they will undergo a security 
investigation. 
Thus, a bona
de contender who proceeds to Stage V is the de 
facto winner. 
Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 8/26
Initial Analysis 
Table : The bidding stages. 
Stage Reilly Contemporary 
Stage II r2=35 c2=35 min 65% 
Stage III r3=25 c3=25 
Stage IV r4=40 c4=40 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 9/26
Fact 1 
Since to pass stage II at least 65% is needed, in terms of points we 
know that: 
35  r2; c2  35  0:65 = 22:75  23: 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 10/26
Assumption 1 
In Stage IV the points are given based on the bid's $-value. 
Assume that the distribution is inversely linear; that is, lowest bid 
(Reilly's $67) is given 40 points, and since Contemporary's bid was 
$81, and 81 = 1:21  67, we assume that Contemporary gets 
40=1:21 = 33 points. 
This assumption is consistent with the explanation in the Request 
for Proposals. 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 11/26
Given that Contemporary obtained more points, and given the 
above, we have the following: 
r2 + r3 + r4  c2 + c3 + c4 
) r2 + r3 + 40  c2 + c3 + 33 
) (c2  r2) + (c3  r3)  7 () 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 12/26
On the other hand we know that jc2  r2j  12, because the 
highest possible score is 35 and the lowest possible score is 23. 
So the largest dierence possible in the scoring in Stage II is 
12 = 35  23. 
So we consider the two possible scenarios under the assumption 
that Contemporary scored higher overall. 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 13/26
Scenario 1 
Suppose Contemporary won Stage II. 
Then c2  r2  12. But then using line (), we have that 
(c2  r2)  7  (c3  r3) 
and so we know that it is impossible to have 7  (c3  r3)  12, 
which in turns means that it is impossible to have r3  c3  5. 
In other words, if Contemporary won the bid, and Contemporary 
won Stage II, then it is impossible for Reilly to win Stage III by 
more than 5 points. 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 14/26
Scenario 2 
Suppose Contemporary lost Stage II. 
Then (c2 r2)  1, so again using (), we get that (c3 r3)  8. 
In other words, if Contemporary won the contest, but 
Contemporary lost Stage II, then Contemporary must have won 
Stage III by at least 8 points. 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 15/26
Initial Conclusion 
When the actual scores showing that Contemporary won are made 
public, we can check that they are consistent as follows: 
If Contemporary won Stage II, then Reilly cannot lead 
Contemporary by more than 5 points in Stage III. 
If Contemporary lost Stage II, then Contemporary must lead Reilly 
in Stage III by at least 8 points. 
If neither is true, then the results are inconsistent. 
Of course, this is the case under Assumption 1, which is reasonable. 
Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 16/26
Final Analysis 
A few days later we were given more information: 
Table : Reilly's scores released. 
Stage Reilly Contemporary 
2 25/35 ? 
3 18/25 ? 
Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 17/26
Plugging those into (), we obtain: 
(c2  25) + (c3  18)  7; 
which means that c2 + c3  50. On the other hand, c2 + c3  60, 
so together we get: 
50  c2 + c3  60: 
Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 18/26
Scenario 2 not possible 
In other words, Contemporary must be within 9 points of the 
maximum possible for both Stage II and Stage III. 
In particular, this means that Scenario 2 (Contemporary lost in 
Stage II) is not possible: 
for suppose that it did lose in Stage II, (c2  r2)  1, so 
c2  1 + 25 = 24. So c2  24, but then Contemporary is not 
within 9 points of the maximum possible. 
So Scenario 2 is not possible. 
Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 19/26
For Contemporary not to win Stage III, it must be the case that 
c3  18, which means that it lost at least 7 = 25  18 points in 
Stage III. 
Thus, if Contemporary lost more than 2 points in Stage II (i.e., 
Contemporary won Stage II, but scored at most 33/35), it must 
have necessarily won Stage III. 
Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 20/26
Final Conclusion 
Therefore, we know that Contemporary must have won Stage II, 
and either won Stage III as well, or lost no more than two points in 
Stage II. 
Overall, Contemporary's combined score in Stage II and Stage III 
must have been very close to the maximum number of points 
possible: 
51  c2 + c3  60: 
What is interesting is that Contemporary can lose Stage III and 
Stage IV, but by scoring nearly perfect in Stage II (i.e., at least 
33/35) it can still win the competition. 
This seems to be an aberration. 
Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 21/26

More Related Content

Viewers also liked

Feasible Combinatorial Matrix Theory - MFCS2013
Feasible Combinatorial Matrix Theory - MFCS2013Feasible Combinatorial Matrix Theory - MFCS2013
Feasible Combinatorial Matrix Theory - MFCS2013Michael Soltys
 
La, permutations, and the hajós calculus - ICALP 2004
La, permutations, and the hajós calculus - ICALP 2004La, permutations, and the hajós calculus - ICALP 2004
La, permutations, and the hajós calculus - ICALP 2004Michael Soltys
 
Boolean Programs and Quantified Propositional Proof System -
Boolean Programs and Quantified Propositional Proof System - Boolean Programs and Quantified Propositional Proof System -
Boolean Programs and Quantified Propositional Proof System - Michael Soltys
 
The proof theoretic strength of the Steinitz exchange theorem - EACA 2006
The proof theoretic strength of the Steinitz exchange theorem - EACA 2006The proof theoretic strength of the Steinitz exchange theorem - EACA 2006
The proof theoretic strength of the Steinitz exchange theorem - EACA 2006Michael Soltys
 
Forced repetitions over alphabet lists
Forced repetitions over alphabet listsForced repetitions over alphabet lists
Forced repetitions over alphabet listsMichael Soltys
 
How Safe is your Data?
How Safe is your Data?How Safe is your Data?
How Safe is your Data?Michael Soltys
 
Circuit Complexity of Shuffle - IWOCA 2013
Circuit Complexity of Shuffle - IWOCA 2013Circuit Complexity of Shuffle - IWOCA 2013
Circuit Complexity of Shuffle - IWOCA 2013Michael Soltys
 
An algorithmic view of Computer Science
An algorithmic view of Computer ScienceAn algorithmic view of Computer Science
An algorithmic view of Computer ScienceMichael Soltys
 
Feasible proofs of matrix identities with csanky's algorithm - CSL 2005
Feasible proofs of matrix identities with csanky's algorithm - CSL 2005Feasible proofs of matrix identities with csanky's algorithm - CSL 2005
Feasible proofs of matrix identities with csanky's algorithm - CSL 2005Michael Soltys
 

Viewers also liked (10)

Algorithms on Strings
Algorithms on StringsAlgorithms on Strings
Algorithms on Strings
 
Feasible Combinatorial Matrix Theory - MFCS2013
Feasible Combinatorial Matrix Theory - MFCS2013Feasible Combinatorial Matrix Theory - MFCS2013
Feasible Combinatorial Matrix Theory - MFCS2013
 
La, permutations, and the hajós calculus - ICALP 2004
La, permutations, and the hajós calculus - ICALP 2004La, permutations, and the hajós calculus - ICALP 2004
La, permutations, and the hajós calculus - ICALP 2004
 
Boolean Programs and Quantified Propositional Proof System -
Boolean Programs and Quantified Propositional Proof System - Boolean Programs and Quantified Propositional Proof System -
Boolean Programs and Quantified Propositional Proof System -
 
The proof theoretic strength of the Steinitz exchange theorem - EACA 2006
The proof theoretic strength of the Steinitz exchange theorem - EACA 2006The proof theoretic strength of the Steinitz exchange theorem - EACA 2006
The proof theoretic strength of the Steinitz exchange theorem - EACA 2006
 
Forced repetitions over alphabet lists
Forced repetitions over alphabet listsForced repetitions over alphabet lists
Forced repetitions over alphabet lists
 
How Safe is your Data?
How Safe is your Data?How Safe is your Data?
How Safe is your Data?
 
Circuit Complexity of Shuffle - IWOCA 2013
Circuit Complexity of Shuffle - IWOCA 2013Circuit Complexity of Shuffle - IWOCA 2013
Circuit Complexity of Shuffle - IWOCA 2013
 
An algorithmic view of Computer Science
An algorithmic view of Computer ScienceAn algorithmic view of Computer Science
An algorithmic view of Computer Science
 
Feasible proofs of matrix identities with csanky's algorithm - CSL 2005
Feasible proofs of matrix identities with csanky's algorithm - CSL 2005Feasible proofs of matrix identities with csanky's algorithm - CSL 2005
Feasible proofs of matrix identities with csanky's algorithm - CSL 2005
 

Recently uploaded

Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...Boston Institute of Analytics
 
How we prevented account sharing with MFA
How we prevented account sharing with MFAHow we prevented account sharing with MFA
How we prevented account sharing with MFAAndrei Kaleshka
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDRafezzaman
 
Heart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis ProjectHeart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis ProjectBoston Institute of Analytics
 
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024thyngster
 
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...GQ Research
 
Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 217djon017
 
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...Thomas Poetter
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryJeremy Anderson
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改yuu sss
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max PrincetonTimothy Spann
 
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...ssuserf63bd7
 
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...Boston Institute of Analytics
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024Timothy Spann
 
detection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxdetection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxAleenaJamil4
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesTimothy Spann
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfBoston Institute of Analytics
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)jennyeacort
 

Recently uploaded (20)

Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
 
How we prevented account sharing with MFA
How we prevented account sharing with MFAHow we prevented account sharing with MFA
How we prevented account sharing with MFA
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
 
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
Heart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis ProjectHeart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis Project
 
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
 
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
Biometric Authentication: The Evolution, Applications, Benefits and Challenge...
 
Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2
 
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data Story
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max Princeton
 
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
 
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
 
detection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxdetection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptx
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
 

Fair ranking in competitive bidding procurement: a case analysis

  • 1. Fair ranking in competitive bidding procurement A case analysis Michael Soltys KES 2014 18th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems IS13: Modelling with Qualitative and Quantitative Pairwise Comparisons Pomorski Park Naukowo-Techniczny (PPNT) Gdynia September 15, 2014 Fair ranking - Soltys KES'14 PPNT Gdynia Title - 1/26
  • 2. Fair and transparent procurement procedures are a cornerstone of a well functioning free-market economy. Competitive bid procurement involves suppliers of services competing for a contract. The traditional method involves the Owner issuing an Invitation, directed to a group of suppliers. Bidders who wish to participate submit binding oers to do the required service by a
  • 3. xed rate and time and in accordance with the Owner's requirements. The Owner then reviews and evaluates all on-time bids and chooses one Bidder to do the work. The Owner and that Bidder then sign a contract to do the work. Fair ranking - Soltys KES'14 PPNT Gdynia Introduction - 2/26
  • 4. General rules of bidding For the Owner, the implied obligations or terms of the Bid Contract will be: 1. A duty of full disclosure 2. A duty of fairness and good faith For the Bidder the implied term of the Bid Contract will be: 1. Honor the rules of the Invitation. Fair ranking - Soltys KES'14 PPNT Gdynia Introduction - 3/26
  • 5. If any of these implied obligations are broken, then the wronged party has grounds for a lawsuit for breach of contract. Note that in the case that we examine in this paper we make no claims whatsoever of a breach of contract. Rather, we claim that the rules put in place to rank the bids did not adequately select the best bid. Fair ranking - Soltys KES'14 PPNT Gdynia Introduction - 4/26
  • 6. Reilly vs Contemporary In our case the Owner is the Ministry of Community Safety and Correctional Services, and we consider two Bidders: Reilly Security, and the second Bidder was Contemporary Security. While Reilly is an Ontario based company, Contemporary Security is a Vancouver-based subsidiary of an American company. All companies (a.k.a., proponents) were bidding for a contract to provide security during the Pan American games to be held in Toronto in 2015. Since the Owner is the Ministry, the party concerned regarding the $-value of the bid are the tax payers of Ontario; indeed, the rules of the bid should satisfy the tax payers that they are getting the most competitive bid that can ful
  • 7. ll the obligations of the contract. Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 5/26
  • 8. The Ministry conducted the evaluation of proposals in the following stages: Stage I consists of a review to determine which proposals comply with all of the mandatory requirements. There were only two proposals left at the end of this stage: Reilly and Contemporary. Stage II consists of a scoring by the Ministry of each quali
  • 9. ed proposal on the basis of the rated criteria. Only proponents meeting a minimum score of 65% will proceed to Stage III. Stage II will be worth 35% of the proponent's overall score. Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 6/26
  • 10. Stage III consists of a scoring by the Ministry of each quali
  • 11. ed proponent meeting the minimum score in Stage II on the basis of a presentation. Stage III will be worth 25% of the proponent's overall score. Stage IV consists of the sealed pricing envelope provided by each proponent meeting the minimum score in Stage II and completing Stage III will then be opened and Stage IV will consist of a scoring of the pricing submitted. Stage IV will be worth 40% of the proponent's overall score. Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 7/26
  • 12. At the conclusion of Stage IV, the scores from Stage II, Stage III, and Stage IV will be added, and the highest scoring proponent will proceed to Stage V| where they will undergo a security investigation. Thus, a bona
  • 13. de contender who proceeds to Stage V is the de facto winner. Fair ranking - Soltys KES'14 PPNT Gdynia Contention - 8/26
  • 14. Initial Analysis Table : The bidding stages. Stage Reilly Contemporary Stage II r2=35 c2=35 min 65% Stage III r3=25 c3=25 Stage IV r4=40 c4=40 Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 9/26
  • 15. Fact 1 Since to pass stage II at least 65% is needed, in terms of points we know that: 35 r2; c2 35 0:65 = 22:75 23: Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 10/26
  • 16. Assumption 1 In Stage IV the points are given based on the bid's $-value. Assume that the distribution is inversely linear; that is, lowest bid (Reilly's $67) is given 40 points, and since Contemporary's bid was $81, and 81 = 1:21 67, we assume that Contemporary gets 40=1:21 = 33 points. This assumption is consistent with the explanation in the Request for Proposals. Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 11/26
  • 17. Given that Contemporary obtained more points, and given the above, we have the following: r2 + r3 + r4 c2 + c3 + c4 ) r2 + r3 + 40 c2 + c3 + 33 ) (c2 r2) + (c3 r3) 7 () Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 12/26
  • 18. On the other hand we know that jc2 r2j 12, because the highest possible score is 35 and the lowest possible score is 23. So the largest dierence possible in the scoring in Stage II is 12 = 35 23. So we consider the two possible scenarios under the assumption that Contemporary scored higher overall. Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 13/26
  • 19. Scenario 1 Suppose Contemporary won Stage II. Then c2 r2 12. But then using line (), we have that (c2 r2) 7 (c3 r3) and so we know that it is impossible to have 7 (c3 r3) 12, which in turns means that it is impossible to have r3 c3 5. In other words, if Contemporary won the bid, and Contemporary won Stage II, then it is impossible for Reilly to win Stage III by more than 5 points. Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 14/26
  • 20. Scenario 2 Suppose Contemporary lost Stage II. Then (c2 r2) 1, so again using (), we get that (c3 r3) 8. In other words, if Contemporary won the contest, but Contemporary lost Stage II, then Contemporary must have won Stage III by at least 8 points. Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 15/26
  • 21. Initial Conclusion When the actual scores showing that Contemporary won are made public, we can check that they are consistent as follows: If Contemporary won Stage II, then Reilly cannot lead Contemporary by more than 5 points in Stage III. If Contemporary lost Stage II, then Contemporary must lead Reilly in Stage III by at least 8 points. If neither is true, then the results are inconsistent. Of course, this is the case under Assumption 1, which is reasonable. Fair ranking - Soltys KES'14 PPNT Gdynia Initial Analysis - 16/26
  • 22. Final Analysis A few days later we were given more information: Table : Reilly's scores released. Stage Reilly Contemporary 2 25/35 ? 3 18/25 ? Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 17/26
  • 23. Plugging those into (), we obtain: (c2 25) + (c3 18) 7; which means that c2 + c3 50. On the other hand, c2 + c3 60, so together we get: 50 c2 + c3 60: Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 18/26
  • 24. Scenario 2 not possible In other words, Contemporary must be within 9 points of the maximum possible for both Stage II and Stage III. In particular, this means that Scenario 2 (Contemporary lost in Stage II) is not possible: for suppose that it did lose in Stage II, (c2 r2) 1, so c2 1 + 25 = 24. So c2 24, but then Contemporary is not within 9 points of the maximum possible. So Scenario 2 is not possible. Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 19/26
  • 25. For Contemporary not to win Stage III, it must be the case that c3 18, which means that it lost at least 7 = 25 18 points in Stage III. Thus, if Contemporary lost more than 2 points in Stage II (i.e., Contemporary won Stage II, but scored at most 33/35), it must have necessarily won Stage III. Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 20/26
  • 26. Final Conclusion Therefore, we know that Contemporary must have won Stage II, and either won Stage III as well, or lost no more than two points in Stage II. Overall, Contemporary's combined score in Stage II and Stage III must have been very close to the maximum number of points possible: 51 c2 + c3 60: What is interesting is that Contemporary can lose Stage III and Stage IV, but by scoring nearly perfect in Stage II (i.e., at least 33/35) it can still win the competition. This seems to be an aberration. Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 21/26
  • 27. The $-value of the bid not given the proper weight Let x = 60 (c2 + c3), that is, x tells us how far Contemporary is from a perfect score in Stage II and Stage III. That is, if x = 0 then they got the maximum possible number of points (c2 + c3 = 60). On the other hand, if x = 9 then they got the lowest possible score and still win (c2 + c3 = 51). Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 22/26
  • 28. We now compute the size of the bid of Contemporary as a function of x. That is, we know that if x = 9 then Contemporary's bid, denoted cbid, is $81 million, and we want to know how high a bid Contemporary could make and still win (the higher cbid, the smaller the x must be): 40 67 cbid = 33 (9 x); which means that cbid = 2680 24 + x . In other words, if Contemporary scored perfect in Stage II and Stage III, i.e., x = 0, then it could bid cbid = 2680=24 = 111 million and still win! Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 23/26
  • 29. 8 6 4 2 0 110 100 90 80 x is points lost in Stage II 3; 0 means no points lost; 9 is max nr. of points that can be lost cbid = 2680/(24 + x) in millions of $’s Fig. 1. Relationship between points lost and the height of the possible winning bid. Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 24/26
  • 30. One more conclusion In the above graph, x counts missing points of Contemporary in Stage II and Stage III. So, for example, if x = 0, then Contemporary is not missing any points (perfect score in Stage II and Stage III) and so Contemporary can win with a bid of $111 million. On the other hand, when x = 9, this is the max number of points that Contemporary can miss in Stage II and Stage III and still win, so in that case it cannot have a bid of more than $81 million. The plot shows the $-values in between. Again, it seems to be an aberration for Contemporary to be able to have a much higher $-value bid than Reilly, and be awarded the contract. Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 25/26
  • 31. Pan Am Toronto 2015 [cited April 4, 2014]. URL http://bit.ly/1dTPd4F R. J. Brennan, NDP wants probe of Pan Am security contract that went to U.S.
  • 32. rm, The Toronto Star. URL http://on.thestar.com/1jFvENj R. J. Brennan, NDP wants $81-million Pan Am security contract under microscope, The Hamilton Spectator. URL http://bit.ly/1jFvc1y R. J. Brennan, Auditor general to audit Pan Am Games security contract, The Toronto Star. URL http://on.thestar.com/1eeyROH S. Kari, Ontario Auditor-General to audit Pan Am contracts, The Globe and Mail. URL http://bit.ly/1jF1N7I R. C. Worthington, Legal obligations for public purchasers (May 2012) [cited April 4, 2014]. URL http://bit.ly/1j9UDra Request for proposals for private security services for the 2015 Pan Am / Parapan Am games, Request for Proposals OPP-0723, Ministry of Community Safety and Correctional Services (September 2013). Fair ranking - Soltys KES'14 PPNT Gdynia Final Analysis - 26/26