Littlefield round 2 report

Module Code: BMGT43830

Date of Submission: 19th November 2021

Group Members:
Student Name Student Number Student Email
Sinead Kirwan 17405026 sinead.kirwan1@ucdconnect.ie

Charu Suri 21201453 charu.suri@ucdconnect.ie

Vishal Shankar 21203417 vishal.shankar@ucdconnect.ie

Akshay Saini 21201320 akshay.saini@ucdconnect.ie

Thomas Dixon 20203504 tom.dixon@ucdconnect.ie

Jose Ben Kallarakkal 21203377 jose.kallarakkal@ucdconnect.ie

Declaration and signatures: -

Vishal Shankar
1. Round 1 recap
The duration of round 1 was 336 simulated days (7 real days). One hour of real time
translated into two simulated days. The assembly process consists of four steps carried out at
three stations called board stuffing (station 1), testing (station 2) and tuning (station 3). In
round one we had control over two decisions. The decisions were which contract and how
many machines were at each station.

Contract- On the day that the simulation became live we changed the current contact to
contract 3 ($650 margin, ½ day lead-time). This decision was made to generate revenue to
enable the purchase of more machines.

Machines- We purchased a machine at station 1 on day 105 for $90,000 bringing up station
1’s machine count to 4. This decision was made since the utilisation was over 60% for
multiple days and it was causing long queues to build up. Once the extra machine was
purchased it reduced the utilisation and in turn reduced the queue time. We purchased a
machine at station 3 on day 192 for $100,000 bringing up station 3’s machine count to 2.

Overall, the team has actively taken part and keenly monitored the queueing time, lead times
and the demand rate. For the majority of the simulation we stayed on contract 3 ($650
margin, ½ day lead-time) as we felt this brought in the most revenue for us each day and for
the most part we were able to match the lead-time.

2. Round 2 action plan recap

We had aimed to begin the simulation on contract 2, (price = $1000; quoted lead time = 1
day; maximum lead time = 2 days). This is where our processing system was best running
prior to the close of the live simulation. If the system was struggling at the beginning of the
live simulation we had aimed to revert to contract 1, (price = $750; quoted lead time = 7
days; maximum lead time = 14 days). This will increase the time to build cash reserves but
will reduce the strain on the system to allow for improved functionality.
In the previous live simulation we had an issue with queuing build up at the testing station. In
this simulation we had wanted to adjust the scheduling policy at station 2. The plan was to
initially allow priority for tuning chips to be processed. This will also in theory prevent a
queuing issue at station 3.

Once the cash reserve had increased, a decision was going to be made to purchase another
machine at station 1 ‘Board Stuffing’ or Station 3 ‘Tuning’. These were two areas that
struggled to meet the high demand of reduced lead time contracts. Once we had rectified the
queuing issue at Station 2 and 3 through the adjusted scheduling policy, an educated decision
was going to be made through our analysis on the appropriate purchase of a machine for
station 1 or 3.

The machine will be purchased either by cash or the finance option of a loan available from
day 100. This was going to depend on how well the process is performing and the size of our
accumulated cash reserve. Taking into consideration the future spend on inventory. A healthy
cash reserve will see the machine being bought without a loan, provided the cost of inventory
can be maintained by a reduced cash reserve and an increased demand in reduced lead times.
Vice versa, if there is an unhealthy reserve a loan could be a better option to maintain cash
reserves for inventory while also ensuring the maintenance of an efficient productive system.

In order to manage the inventory efficiently, that is to have safety stock at any point and
avoid stock outs or overstocks, we also calculated the reorder point after determining the
reorder level on the basis of demand and reorder quantities in every 20 days. By looking at
the trend of daily demand we were able to infer further demand in the upcoming days and by
taking an average helped us to determine the reorder quantity. Keeping frequent checks on
inventory will help us run the operations smoothly.
 EOQ= SQRT((2*Demand*Co/Ch))

Demand 12.22*365*60= 267618

Ordering Cost $1000

Unit Cost $10

Holding Cost (assuming) 10% of unit cost (10*0.1)= $1

EOQ SQRT(2*267618*1000/1)= 23135.17

Batches 23135.17/60= 385.59

Basic ROL= L*Avg Demand

Lead Time 4 days

Average Demand 12.22

ROL 12.22*60*4= 2932.80

3. Decisions taken
For the round 2 of simulation we were having the below decisions in hand -
● Reorder point
● Reorder quantity
● No. of machines
● Contract
● Loan
When we got the control of the factory, our machines at all the stations were working at low capacity and
we didn’t see a need to buy another machine. Initially when we weren’t sure how the demand would
fluctuate. We stayed at contract 1, after 20 simulated days we calculated the EOQ and ROL but it came
exactly the same that was automatically the same at the start of simulation which is why we continued
with the already set inventory replenishment levels. With the given set up our machines worked fine for
341 days when we decided to move the contract to contract 2 and the machines at station 1 were utilised
at 100 percent capacity, then we bought another machine at station 1 on the day 351.
Our factory ran smoothly till the end with 4 machines at station 1, 2 machines at station 2, 1 machine at
station 3 and we ultimately ended the simulation with a cash balance of 1,056,744.

Actions that should have been taken but were not:

● Instead of waiting, we should have bought a machine at station 3 as soon as the line
started to lengthen.

On day 345, as the queue at station 3 began to grow, we deliberated for two more hours to see
if the backlog dwindled but machine utilization remained at 100%, and even whether we
should invest in another machine, all while waiting for new data and generally not acting
decisively. We should have bought the machine as soon as we saw the wait start to expand,
given that we hadn't yet reached the halfway point and demand was projected to continue to
rise. This wasted chance cost us in the long run because the queue continued to build while
we waited to buy the equipment. We stayed on contract 1, after 20 simulation days and so the
demand wasn't that great, we calculated the EOQ and ROL, and the line eventually thinned
out. Then we changed the contract from 1 to 2 and purchased the machine at station 1
because there was more que at station 1. Later, as demand grew, the situation deteriorated;
the queue at machine 1 was managed, but station 3 was once again overburdened. New
capacity was filling "stale," non-profitable orders, causing all new orders to become "stale,"
and we missed out on money for several days. We had only a few hours in actual time to
make the decision to buy the machine at station 3, which we failed to do, and we should have
made the decision to buy the machine at station 3 when we first saw the line.
 The identical thing was happening at station 1 while we were concentrating on this. We were distracted
by the challenging scenario at station 3 and didn't realize it for another day, but we didn't spend nearly as
much time deciding about this purchase. However, we later purchased the machine at Station 1, which
assisted us in reducing the queue at Station.

4. Future improvements
In the future, the main focus in the Littlefield Simulation should be on Optimal Strategy, and the
results can be used to update the model if necessary. When the game's parameters change, the
reorder level, reorder quantity, and number of machines to be purchased will all update
automatically. The model can also account for the mechanism for calculating the holding cost of
raw materials in inventory, which was previously overlooked.

To summarize, our model can be used to simulate in any industry in order to compete with the firms
of competitors.
Appendix: