Fixtures Status¶

Run this example in Colab¶

This example plots the number of fixtures per month, grouped by status for Very Large Crude Carriers (VLCCs)

Setup

Install the Signal Ocean SDK:

In [ ]:

pip install signal-ocean

pip install signal-ocean

Import required libraries / APIs / packages:

Import Voyages Market Data API and some additional libraries

In [1]:

import pandas as pd
import plotly.express as px
from datetime import date, timedelta, datetime

from signal_ocean import Connection
from signal_ocean.voyages_market_data import VoyagesMarketDataAPI

import pandas as pd import plotly.express as px from datetime import date, timedelta, datetime from signal_ocean import Connection from signal_ocean.voyages_market_data import VoyagesMarketDataAPI

Fill in your personal SignalOcean API subscription key acquired here and open a connection:

In [2]:

signal_ocean_api_key = '' #replace with your subscription key

signal_ocean_api_key = '' #replace with your subscription key

Create a connection

In [3]:

connection = Connection(signal_ocean_api_key)
api = VoyagesMarketDataAPI(connection)

connection = Connection(signal_ocean_api_key) api = VoyagesMarketDataAPI(connection)

Call the API to Get market data

In [4]:

result = api.get_voyage_market_data(vessel_class_id=84, include_fixtures = True, include_matched_fixture = True, include_labels = True)

result = api.get_voyage_market_data(vessel_class_id=84, include_fixtures = True, include_matched_fixture = True, include_labels = True)

Get data into a dataframe

In [5]:

voyages_market_data = pd.DataFrame([r.__dict__ for r in result])

voyages_market_data = pd.DataFrame([r.__dict__ for r in result])

Extract date info from matched fixture

In [6]:

voyages_market_data = voyages_market_data[voyages_market_data['matched_fixture'].notna()]
voyages_market_data['Date'] = voyages_market_data['matched_fixture'].apply(lambda mf: mf.laycan_from)

voyages_market_data = voyages_market_data[voyages_market_data['matched_fixture'].notna()] voyages_market_data['Date'] = voyages_market_data['matched_fixture'].apply(lambda mf: mf.laycan_from)

Transform Data

Drop rows that have no date

In [7]:

voyages_market_data = voyages_market_data[voyages_market_data.Date.map(lambda d: isinstance(d,datetime))]

voyages_market_data = voyages_market_data[voyages_market_data.Date.map(lambda d: isinstance(d,datetime))]

Expand the dataframe to have one row per fixture and extract fixture_status

In [8]:

voyages_market_data=voyages_market_data.explode('fixtures').reset_index(drop=True)
voyages_market_data['fixture_status'] = voyages_market_data['fixtures'].apply(lambda f: f.fixture_status)

voyages_market_data=voyages_market_data.explode('fixtures').reset_index(drop=True) voyages_market_data['fixture_status'] = voyages_market_data['fixtures'].apply(lambda f: f.fixture_status)

Keep only rows with fixture_status info

In [9]:

voyages_market_data = voyages_market_data[voyages_market_data['fixture_status'].notna()][['Date', 'fixture_status', 'voyage_id']]

voyages_market_data = voyages_market_data[voyages_market_data['fixture_status'].notna()][['Date', 'fixture_status', 'voyage_id']]

Keep one fixture report for each status per voyage

In [10]:

voyages_market_data = voyages_market_data.drop_duplicates(subset=['voyage_id', 'fixture_status'], keep='last')
voyages_market_data.head()

voyages_market_data = voyages_market_data.drop_duplicates(subset=['voyage_id', 'fixture_status'], keep='last') voyages_market_data.head()

Out[10]:

	Date	fixture_status	voyage_id
62	2011-07-09 00:00:00+00:00	FullyFixed	I8818E7VEC5455F00
67	2011-08-28 00:00:00+00:00	FullyFixed	I8818E7VEC5BC0600
74	2011-10-10 00:00:00+00:00	FullyFixed	I8818E7VEC60B2000
77	2012-01-25 00:00:00+00:00	OnSubs	I8818E7VEC681C700
81	2012-03-05 00:00:00+00:00	FullyFixed	I8818E7VEC6D0E100

Keep only Month and year info on Date

In [11]:

voyages_market_data['Date'] = pd.to_datetime(voyages_market_data['Date'], utc=True).dt.to_period('M')
voyages_market_data.head()

voyages_market_data['Date'] = pd.to_datetime(voyages_market_data['Date'], utc=True).dt.to_period('M') voyages_market_data.head()

C:\Users\l.kolonis\Anaconda3\envs\python38\lib\site-packages\pandas\core\arrays\datetimes.py:1143: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.
  warnings.warn(

Out[11]:

	Date	fixture_status	voyage_id
62	2011-07	FullyFixed	I8818E7VEC5455F00
67	2011-08	FullyFixed	I8818E7VEC5BC0600
74	2011-10	FullyFixed	I8818E7VEC60B2000
77	2012-01	OnSubs	I8818E7VEC681C700
81	2012-03	FullyFixed	I8818E7VEC6D0E100

Group and plot

In [12]:

voyages_market_data_grouped = voyages_market_data.groupby(['Date', 'fixture_status']).size().reset_index(name="Count")
voyages_market_data_grouped.sort_values(['Date'], ascending = True, inplace = True)
voyages_market_data_grouped.tail()

voyages_market_data_grouped = voyages_market_data.groupby(['Date', 'fixture_status']).size().reset_index(name="Count") voyages_market_data_grouped.sort_values(['Date'], ascending = True, inplace = True) voyages_market_data_grouped.tail()

Out[12]:

	Date	fixture_status	Count
467	2022-08	FullyFixed	58
466	2022-08	Failed	5
468	2022-08	OnSubs	110
469	2022-09	FullyFixed	3
470	2022-09	OnSubs	7

In [13]:

voyages_market_data_grouped=voyages_market_data_grouped[voyages_market_data_grouped['Date']>='2020-01']
voyages_market_data_grouped['Date'] = voyages_market_data_grouped['Date'].astype(str)

voyages_market_data_grouped=voyages_market_data_grouped[voyages_market_data_grouped['Date']>='2020-01'] voyages_market_data_grouped['Date'] = voyages_market_data_grouped['Date'].astype(str)

In [14]:

fig = px.bar(voyages_market_data_grouped, x="Date", y="Count", color='fixture_status', orientation='v',
             barmode='group', height=500, width=1000,
             title='VLCC Fixture Status over time',
             color_discrete_sequence=px.colors.qualitative.Vivid)
fig.show()

fig = px.bar(voyages_market_data_grouped, x="Date", y="Count", color='fixture_status', orientation='v', barmode='group', height=500, width=1000, title='VLCC Fixture Status over time', color_discrete_sequence=px.colors.qualitative.Vivid) fig.show()

In [ ]: