Decision Mathematics 1

Practice Paper 4

Instructions
 Use black ink or ball-point pen.
 If pencil is used for diagrams/sketches/graphs it must be dark (HB or B).
 Fill in the boxes at the top of this page with your name, centre number and candidate number.
 Answer all the questions and ensure that your answers to parts of questions are clearly labelled.
 Answer the questions in the spaces provided – there may be more space than you need.
 You should show sufficient working to make your methods clear. Answers without working may
not gain full credit.
 Inexact answers should be given to three significant figures unless otherwise stated.

Information
 A booklet ‘Mathematical Formulae and Statistical Tables’ is provided.
 There are 6 questions in this question paper. The total mark for this paper is 75.
 The marks for each question are shown in brackets – use this as a guide as to how much time to
spend on each question.
 Calculators must not be used for questions marked with a * sign.

Advice
 Read each question carefully before you start to answer it.
 Try to answer every question.
 Check your answers if you have time at the end.
 If you change your mind about an answer, cross it out and put your new answer and any working
underneath.
1.
30 11 21 53 50 39 16 4 60 43

The numbers in the list above represent the lengths, in cm, of some pieces of electrical wire.
The wire is sold in one metre lengths.
(a) Use the first-fit bin packing algorithm to determine how these pieces could be cut from
one metre lengths. You should ignore wastage due to cutting.
(3)

The list of numbers above is to be sorted into ascending order.

Starting at the left-hand end of the list, after three passes of the bubble sort, the list is
11 21 30 16 4 39 43 50 53 60

(b) (i) Write down the list that results at the end of the fourth pass.
(ii) Write down the number of comparisons and swaps performed during the fourth pass.

(3)

The original list of numbers is now to be sorted into descending order.

(c) Perform a quick sort to obtain the sorted list. You should show the result of each pass
and identify your pivots clearly.
(4)
(d) Use the first-fit decreasing bin packing algorithm to determine how these pieces could
be cut from one metre lengths. You should ignore wastage due to cutting.
(3)

(Total 13 marks)
Mark scheme for Question 1
Examiner comment
2. (a) Explain the difference between the classical travelling salesperson problem and the
practical travelling salesperson problem.
(2)
A B C D E F G
A – 31 15 12 24 17 22
B 31 – 20 25 14 25 50
C 15 20 – 16 24 19 21
D 12 25 16 – 21 32 17
E 24 14 24 21 – 28 41
F 17 25 19 32 28 – 25
G 22 50 21 17 41 25 –

The table above shows the least direct distances, in miles, between seven towns, A, B, C, D,
E, F and G. Yiyi needs to visit each town, starting and finishing at A, and wishes to minimise
the total distance she will travel.

(b) Show that there are two nearest neighbour routes that start from A. State these routes and
their lengths.
(3)
(c) Starting by deleting A, and all of its arcs, find a lower bound for the length of Yiyi’s
route.

(3)
(d) Use your results to write down the smallest interval which you can be confident contains
the optimal length of Yiyi’s route.
(2)
(Total 10 marks)

Mark scheme for Question 2

Examiner comment
3. (a) In the context of the critical path analysis, define the term ‘total float’.
(2)

Figure 3

Figure 3 is the activity network for a building project. The activities are represented by the
arcs. The number in brackets on each arc gives the time, in days, to complete the activity.
Each activity requires exactly one worker. The project is to be completed in the shortest
possible time.

(b) Complete Diagram 1 in the answer book to show the early event times and the late event
times.
(3)
(c) State the critical activities.
(1)
(d) Calculate the maximum number of days by which activity G could be delayed without
affecting the shortest possible completion time of the project. You must make the
numbers used in your calculation clear.
(2)
(e) Calculate a lower bound for the number of workers needed to complete the project in the
minimum time. You must show your working.
(2)
Question 3 continued

The project is to be completed in the minimum time using as few workers as possible.

(f) Schedule the activities using Grid 1 in the answer book.

(4)

(Total 14 marks)
Mark scheme for Question 3
Examiner comment
4. The tableau below is the initial tableau for a three-variable linear programming problem in x,
y and z. The objective is to maximise the profit, P.

Basic variable x y z r s t Value

r 15 –2 3 1 0 0 180
s 10 1 1 0 1 0 80
t 1 6 –2 0 0 1 100
P –1 –2 –5 0 0 0 0

(a) Using the information in the tableau, write down

(i) the objective function,
(ii) the three constraints as inequalities.
(3)
(b) Taking the most negative number in the profit row to indicate the pivot column at each
stage, solve this linear programming problem. Make your method clear by stating the
row operations you use.
(8)
(c) State the final values of the objective function and each variable.
(2)

(Total 13 marks)

Mark scheme for Question 4

Examiner comment
5. A linear programming problem in x and y is described as follows.
Minimise C = 2x + 3y

subject to
x+y8
x<8
4y  x
3y  9 + 2x

(a) Add lines and shading to Diagram 1 in the answer book to represent these constraints.
(4)

(b) Hence determine the feasible region and label it R.

(1)

(c) Use the objective line (ruler) method to find the exact coordinates of the optimal vertex,
V, of the feasible region. You must draw and label your objective line clearly.
(3)

(d) Calculate the corresponding value of C at V.

(1)

The objective is now to maximise 2x + 3y, where x and y are integers.

(e) Write down the optimal values of x and y and the corresponding maximum value
of 2x + 3y.
(2)

A further constraint, y  kx, where k is a positive constant, is added to the linear

programming problem.

(f) Determine the least value of k for which this additional constraint does not affect the
feasible region.
(2)

(Total 13 marks)
Mark scheme for Question 5
Examiner comment
6.

[The weight of the network is 20x + 17]

(a) Explain why it is not possible to draw a network with an odd number of vertices of odd
valency.
(2)

Figure 3 represents a network of 12 roads in a city. The expression on each arc gives the time,
in minutes, to travel along the corresponding road.
(b) During rush hour one day x = 9
(i) Starting at A, use Prim’s algorithm to find a minimum spanning tree. You must state
the order in which you select the arcs of your tree.
(ii) Calculate the weight of the minimum spanning tree.
(4)
You are now given that x > 3
A route that minimises the total time taken to traverse each road at least once needs to be
found. The route must start and finish at the same vertex.
The route inspection algorithm is applied to the network in Figure 3 and the time taken for
the route is 162 minutes.
(c) Determine the value of x, showing your working clearly.
(6)

(Total 12 marks)

Mark scheme for Question 6

Examiner comment

TOTAL FOR PAPER: 75 MARKS

 A level Further Mathematics – Decision Mathematics 1 –
Practice Paper 04 – Mark scheme –
Mark scheme for Question 1 (Examiner comment) (Return to Question 1)
Mark
Question Scheme
s
1(a) M1
A1

M1

(3)
(b)(i)
B1
11 21 16 4 30 39 43 50 53 60
(ii) Comparisons: 6 Swaps: 2 B1B1
(3)
(c)
M1
A1
A1ft
A1

(4)
(d) M1
A1

M1
(3)

(13 marks)
 Mark scheme for Question 2 (Examiner comment) (Return to Question 2)

Question Scheme Marks

2(a) e.g. in the practical problem each vertex must be visited at least once. In
B2, 1, 0
the classical problem each vertex must be visited just once
(2)
(b)
A–D–C–F–B–E–G–A M1
12+16+19+25+14+41+22 = 149
A1

A–D–C–F–G–E–B–A
A1
12+16+19+25+41+14+31 = 158
(3)
(c) RMST weight = 86 (miles) B1
86 + 12 + 15 = 113 (miles) M1A1
(3)
(d) 113 optimal distance 149 B2, 1, 0
(2)
(10 marks)
 Mark scheme for Question 3 (Examiner comment) (Return to Question 3)

Question Scheme Marks

3(a) The total float F (i, j) of activity (i , j) is defined to be

F ( i , j )=l j−e i−duration ( i , j ), where e iis the earliest time for event i and l j M1A1
is the latest time for event j (see note below)

(2)
(b)

M1

A1

A1

(3)
(c) Critical activities: A C J M B1
(1)
(d) G can be delayed by 21 – 11 – 3 = 7 (days) M1A1
(2)
(e)
so lower bound is 3 workers M1A1

(2)
 Mark
Question Scheme
s
3(f)

M1
A1
A1
A1

(4)
(14 marks)
 Mark scheme for Question 4 (Examiner comment) (Return to Question 4)

Mark
Question Scheme
s

4(a)(i) P = x + 2y + 5z B1
(ii) 15x − 2y + 3z 180
10x + y + z 80 M1A1
x + 6y − 2z 100
(3)
(b)

M1A1
M1A1

M1
A1ft

M1A1

(8)
(c) P = 364; x = 0; y =12; z = 68; r = s= 0; t = 164 M1A1
(2)

(13 marks)
Mark scheme for Question 5 (Examiner comment) (Return to Question 5)
 Mark
Question Scheme
s

5(a)

B1

B1

B1
B1
(4)
(b) Correct R labelled B1

(1)
(c) Objective line drawn B1

M1
(oe) dA1

(3)
(d)
B1
(oe)
(1)
 Mark
Question Scheme
s

5(e) (7,7) B1
35 B1

(2)
(f)
M1A1
y
(2)

(13 marks)
 Mark scheme for Question 6 (Examiner comment) (Return to Question 6)

Question Scheme Marks

6(a) e.g. (each arc contributes 1 to the orders of two nodes, and so) the sum
of the orders of all the nodes is equal to twice the number of arcs
Which implies that the sum of the orders of all the nodes is even and B2, 1, 10
therefore there must be an even (or zero) number of vertices of odd
order hence there cannot be an odd number of vertices of odd order
(2)
(b)(i) Prim: AB, AD, BC, DE; EF, FG; GJ, HJ M1 A1 A1
(ii) Weight = 92 (mins) B1

(4)
(c) 20x + 17 + repeart arcs = 162
B1
B and E are the only two odd nodes so must be apired
4x + 3 (BE) is clearly greater in value than x + (2x + 3) + x – 1 = 4x + 2
(AB, AD, DE) so repeated arcs are either AB, AD, DE (4x + 2) or BC, B1
CE (7x – 17)
4x + 2 = 7x – 17 19/3 B1

x < 19/3 20x + 7 – 17 = 145 ⸫ x = 6 M1A1

x < 19/3 20x + 4x +2 = 145 ⸫ x = 143/24 < 19/3 so x ≠ 143/24 A1

(6)

(12 marks)
 A level Further Mathematics – Decision Mathematics 1 –
Practice Paper 04 – Examiner report –
Examiner comment for Question 1 (Mark scheme) (Return to Question 1)
1. Examiners reported that a small number of students struggled in applying the first-fit bin
packing algorithm in part (a). This was mainly down to not applying the algorithm correctly.
First fit is just that; students must decide if the current item under consideration will fit in the
first bin rather than the most recent bin used. In this part a number of students placed the 39
in the third bin (and not the second bin) and others did not place the 4 in the first bin.
Full marks in part (b) was rare. Some students completed the full sort rather than stopping
after the fourth pass. Others completed four passes on the provided list so effectively
obtaining the seventh pass. For those who did carry out the correct number of passes or who
made their fourth pass clear, errors were rare. In part (ii) of (b), only the more able students
were able to correctly establish that 6 comparisons were needed with most stating that
9 comparisons were required. This demonstrated something of a lack of understanding of
how the Bubble Sort works. Deducing that 2 swaps took place seemed more straightforward
for most students.
Many correct solutions were seen in part (c), but a number of students did not choose their
pivots consistently, switching between middle-left and middle-right pivots during the course
of the quick sort algorithm. A number of students either lost an item or changed an item
during the sort, and in a small number of cases only one pivot was chosen per iteration. As
stated in previous examiners’ reports, in cases such as this in which the list appears to be in
the correct order after three passes a fourth pass (pivoting on the 4) is required to successfully
complete the algorithm. Students should be reminded that items should remain in the order
from the previous pass as they move into sub-lists. Pivots were usually chosen consistently
although the spacing and notation on some solutions made these difficult for examiners to
follow. Some students over complicated the process by insisting on using a different ‘symbol’
to indicate the pivots for each pass. Those students who sorted into ascending order usually
remembered to reverse their list at the end to gain full credit although a number of students
left their list in ascending order.
The first-fit decreasing in part (d) was well carried out with only a small minority failing to
attempt this part. There were a large number of wholly correct answers. A small number
performed first-fit increasing therefore scoring no marks. A small minority of students lost all
three marks by placing the 43 in the 3rd rather than 2nd bin (so failing to apply the algorithm
at its first real test). Some students wrote totals in the bin rather than the next value. A variety
of different layouts were used but in nearly all cases were easy to read and decipher.
 Examiner comment for Question 2 (Mark scheme) (Return to Question 2)
2. For part (a), many candidates could accurately describe the difference between the practical
and classical problem, with accurate and correct terminology seen in most responses.
However, this initial question differentiated between those that had a secure grasp of the topic
and those that were much less secure in their knowledge, with a significant number referring
to ‘arcs needing to be traversed’, rather than the need to visit every vertex. Those that only
scored partial marks on this part had the correct general idea without being able to pinpoint
the need to visit every vertex.
Whilst part (b) was well answered by the majority of candidates, the most common error was
to fail to return to the starting vertex. There were also a significant number of candidates who
found an upper bound by doubling the length of a minimum spanning tree, so gave answers
of 298 and 316, instead of finding nearest neighbour routes.
Part (c) was also well answered by most candidates, with many securing full marks for an
answer of 113. A significant number of candidates did not find the correct RMST (found
from deleting vertex A and all arcs incident to this vertex), but still managed to secure a mark
for adding the correct two least weighted arcs.
Most candidates scored at least one mark in part (d) on the follow through from an earlier
stated upper bound found in part (b). It was common, however, to lose the final mark in this
part for either the use of a strict inequality for the upper limit, or because of inaccurate earlier
working. It was rare but nonetheless worrying to see statements along the lines of

Examiner comment for Question 3 (Mark scheme) (Return to Question 3)

3. This question also discriminated well leading to a good spread of marks. The modal mark
was 12, only 4.8% scored full marks and 29.5% scored 6 marks or fewer.
Part (a) was poorly answered on the whole. Many definitions of total float muddled the terms
‘events’ and ‘activities’. A large majority of candidates gave incomplete answers, for
example, not clearly stating earliest start time or latest finish time (either saying start and
finish time or early and late times). Some candidates tried to state the mathematical formula
but did not fully explain all the symbols they used. The majority of candidates understood the
idea that it was the time an activity could be delayed for but in many cases, candidates
thought it was the total of the floats for all activities in a network.
In part (b) the dummy between events 6 and 7 caused the most difficulty, with a number of
candidates going via K and getting a late event time of 25 rather than the correct value of 21.
Most could gain the first mark by having only one rogue value and the early event times were
largely correct. In general more mistakes were seen in the bottom half of the boxes.
Candidates should be advised to take time checking their values as a significant number of
subsequent marks can be lost if errors are made in this part.
The majority of students successfully identified the correct critical activities in part (c).
In part (d) the majority of candidates knew the method for the float calculation and showed it
clearly. Others just gave the calculation of 21 – 14 which could have been the latest finish
time – the earliest start time for the event between G and L rather than the correct calculation
21 – 11 – 3. Finding the lower bound in part (e) had more variable success; some did not do a
calculation and tried to argue for a lower bound based on scheduling despite the question
asking for a calculation. Others made either basic arithmetical errors or conceptual errors (the
 most common being calculating the ratio of the earliest possible finish time (30) to the
number of activities (13)) in their calculation.
In part (f) quite a few candidates drew a Gantt chart instead of a scheduling diagram, and so
scored no marks. There were also quite a few instances where this part was left blank. Those
that did schedule tended to make errors on activity H, which needed to take place after
activity D. There were also minor errors in duration lengths seen meaning few scored full
marks in this part. It would be advisable for candidates to check their working carefully to
ensure that preceding activities are completed and that activities do not start before their
earliest start time or continue beyond their latest finish time. Also it was common for at least
one activity to be missing from the scheduling diagram.
Examiner comment for Question 4 (Mark scheme) (Return to Question 4)
4. In Part (a) was answered well by the majority of students with many correctly stating the
objective function and the constraints as inequalities. The most common errors were:
 Stating the objective as either P = −x −2y – 5z or P = x + 2y + 5z = 0,
 stating the constraints as either
15x − 2y + 3z < 180, 15x − 2y + 3z 180, 15x − 2y + 3z + r = 180 or even
15x − 2y + 3z + r  180.
In part (b), many students were able to correctly identify the pivot row and divide and replace
r with z as a basic variable. Most students defined row operations in terms of the new row 1.
There were of course a number of students who made errors in the subsequent row operation
calculations. A significant number stopped after one iteration, in some cases stating that the
solution was not optimal due to negative values in the profit row despite the fact that the
question had asked for the problem to be solved. Of those who did proceed into the second
iteration very few picked negative pivots and many were able to proceed correctly albeit with
some errors in the pivot row calculations.
In part (c), it was clear that students either did not read the question carefully or assumed it
was asking for students to write down the profit equation from the optimal tableau. Therefore
many students did not state P = 364 but instead stated that P + 40x + 0.6r + 3.2s = 364.
Furthermore, many did not state the final values of the slack variables and examiners noted
that it was quite common to see x = 40 rather than x = 0 stated as one of the final values.
 Examiner comment for Question 5 (Mark scheme) (Return to Question 5)
5. This was the most challenging question on the paper for many candidates, with very few
scoring full marks. Most candidates were able to draw the required lines correctly in part (a)
although some were unable to draw lines sufficiently accurately (some drew lines without a
ruler) or sufficiently long enough. The following general principle should always be adopted
by candidates:
 lines should always be drawn which cover the entire graph paper supplied in the answer
book and therefore,
 lines with negative gradient should always be drawn from axis to axis.

The rationale behind this is that until all the lines are drawn (and shaded accordingly) it is
unclear which lines (or parts of lines) will define the boundary of the feasible region. If
candidates only draw the line segments that they believe define the boundary of the feasible
region then examiners are unaware of the order in which the lines were drawn and therefore it
is unclear to examiners why some parts of the lines have been omitted.

In general the lines x  y 8 and 3 y 9  2 x were correctly drawn. One of the most common
errors was with the line 4 y x (many candidates drew y 4 x ), the second most common error
was not showing the line x 8 as distinct from the other three in anyway (many candidates
either ignored or did not notice the strict inequality constraint x  8 ). In part (b) a significant
number of candidates were unable to select the correct feasible region.
In part (c), the majority of candidates drew the correct objective line, however, a line with
reciprocal gradient was often seen or, in a number of cases, no objective line was drawn (and
therefore no marks could be awarded in this part). Some used obscure constant values to plot
the objective line. Some candidates gave an estimate of the optimal vertex using a reading
from their graph, rather than solving the relevant equations simultaneously.
The majority of candidates who correctly answered part (c) usually went on to score the mark
in part (d).
Parts (e) and (f) were often not attempted and many who did attempt part (e) identified the
maximum point as (8, 8) (even though this was not in the feasible region) or gave a point
with non-integer values despite the question explicitly asking for integer values. A minority
of candidates identified the point (7,7) but failed to substitute the values into the objective
function.
Part (f) differentiated well between candidates. Once again the management of the inequality
provided a source of errors for those who had an understanding of what was required in this
part but gave the incorrect answer of k = 1/4.
Examiner comment for Question 6 (Mark scheme) (Return to Question 6)
6. No examiner report available.